Guru3D.com Forums

Go Back   Guru3D.com Forums > General > Links
Links Archived stickys.



Closed Thread
 
Thread Tools Display Modes
Guide to RAM
Old
  (#1)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default Guide to RAM - 07-03-2004, 17:04 | posts: 15,302 | Location: UK

Guide to RAM [Copyright 2005-2007 WildStyle - Guru3D.com]




*******************************************

Brief Intro To RAM Before We Move On
Why is memory so important? You know ... I can't remember; waah what a lame joke :: wipes a tear of joy from his eye ::

To understand this we'll start off with the ordinary A-B-C of memory. Random Access Memory, or volatile memory, is used by the system to store data for processing by a computer's central processing unit (CPU), also known as the processor. RAM stores the data in memory cells that are arranged in grids much like the cells are arranged in a spreadsheet, from which data, in the binary form of 1's and 0's, can be accessed and transferred at random to the processor for processing by the system's software. The faster that process goes up-to a certain threshold, the better your performance can be. When the computer is turned off, all of the data in the RAM memory is lost, hence its alternative name of volatile memory

Up to a point, adding RAM (random access memory) will normally cause your computer to feel faster on certain types of operations. Some applications (think Photoshop or most movie editing and animation packages) need bags of RAM to do their job. If you run them on a PC with too little RAM, they swap constantly and run very slowly. You can get a massive speed boost by adding enough RAM to eradicate the swapping. Programs like these seriously can run 10 to 50 times faster once they have enough RAM. This is no joke.

How does RAM work ? Well, the essence is actually very simple. To store a 1 in the memory cell, a capacitor is filled with electrons. To store a 0, it is emptied .. that's it .. it's that simple. The dilemma with the capacitor's container is that it has a leak. In a matter of a few milliseconds a full bucket becomes empty. Therefore, for memory to work, either the CPU or the memory controller has to come along and recharge all of the capacitors holding a 1 before they discharge. To do this, the memory controller will read the memory and then writes it right back. This refresh operation happens automatically thousands of times per second.

(Source: Hilbert)

*******************************************

Each DIMM (Dual Inline Memory Module - Called dual because it has separate signals to each side of the circuit board) is comprised of three primary components: IC, PCB, and SPD.

An IC (Integrated Circuit) is the formal name for a computer chip. Here, the term is used to describe the small chips on the side(s) of your RAM module.

There are two main types, or form factors of IC.
There is TSOP (Thin Small-Outline Package) and also BGA (Ball Grid Array).

TSOP is surface-mounted to the PCB (the pins are soldered directly to the surface of the PCB) and has gull wing leads (small wires) protruding out the side.



BGA has small solder balls that are at the bottom surface of the chip and no longer visible from the front after the chips are mounted on the PCB.



There are only a handful of companies that actually produce the ICs. Winbond, Samsung, Hynix, Micron and Infineon to name some of the better known and most common (Though relabelling will often occur.)
This is why knowing what IC the RAM has is much more important in determining what it is capable of than going on brand alone. Brand X does not overclock better than brand Y just because it is brand X.

If a good IC is being mass produced, and cheaply to boot, then you may well find this IC used on lower rated, or "value" esque modules too, as well as the premium stuff.

A PCB (Printed Circuit Board) is the thin plate on which chips and other electronic components are placed. More specifically, here the PCB is what the ICs are situated upon.
PCBs are built using multiple layers of metal traces separated by individual sheets of resin, and each layer has its own dedicated set of functions.

The SPD (Serial Presence Detect) is a small EEPROM that contains the DIMMís default configuration such as timings, module size and frequency. This info is extracted by the motherboard chipset. If you have not manually configured the aforementioned configuration settings for your RAM, then your board will pick up the info from the SPD and use it.

----- ----- ----- ----- ----- ----- ----- ----- ----- -----
----- ----- ----- ----- ----- ----- ----- ----- ----- -----

What Does A Memory Manufacturer Do?
The memory module manufacturer can buy the memory chips as a final product from a memory manufacturer like Samsung, Hynix, Infineon, etc; can buy them untested (a.k.a. UTT chips) and test (usually for speed grade) and sort them in-house; or can buy the memory wafer, cut the wafer and pack the integrated circuits by themselves.

The memory module manufacturing process is basically the same for all memory module manufacturers:

1) Apply solder paste to the memory PCB.

2) Put the components on the PCB using a technique called SMT, Surface Mount Technology. This process is also known as pick-and-place.

3) Send the modules inside an oven, where the solder paste will melt, thus soldering the components.

4) Visual inspection.

5) Remove the memory modules from their panels (before this process the memory modules are stuck together in a panel, each panel holds five or six memory modules), a process also known as depanelization.

6) SPD programming and quick manual testing.

7) Memory module testing.

8) Functional testing.

9) Heatsink is attached to the module (if applicable).

10) Labeling.

11) Packing.

12) Shipping to customers.

(Source: Hardware Secrets)

----- ----- ----- ----- ----- ----- ----- ----- ----- -----
----- ----- ----- ----- ----- ----- ----- ----- ----- -----

JEDEC / PC3500/3700/4000+
JEDEC (Joint Electron Device Engineering Council) are the semiconductor engineering standardisation body. They set the standards for the RAM we use. Everything manufactured should conform to JEDEC standards. However, with the highest rating for DDR1 a measly 400MHz (PC3200), RAM companies have taken it upon themselves to go above and beyond the specifications laid out, whilst still retaining backwards compatibility to the appropriate JEDEC standards. In other words, PC4000 and such will work fine on your board, and will help your overclocking by allowing you the necessary headroom to increase the FSB without the RAM becoming the limiting factor. If you do not overclock, you can always downclock the RAM to match your FSB and run 1:1. By doing that, you will normally be able to tighten the latencies too, since the RAM would be running at a lower frequency than what the SPD timings and MHz would suggest.

ECC and Non-ECC
ECC stands for "error correcting code," and it detects and reports memory errors. However, while non-ECC can only detect errors, ECC can actually correct errors without interrupting the other operations of your system.
There is a slight decrease in performance because registers delay memory information for one clock cycle to ensure all communication from the chipset is collected by the clock edge, providing a controlled delay on heavily loaded memories. (Source: Crucial.com) This delay amounts to approximately 3-4%.
ECC is more expensive than Non-ECC, and not all motherboards support it. It's often used in servers rather than desktop systems which generally use Non-ECC.

Buffered and Unbuffered
Buffered Memory = A buffer isolates the memory from the controller to minimize the load on the chipset. It is typically used when the system has a high density of memory and/or when a system has more than 3 memory module sockets.

Unbuffered memory = This is where the chipset controller deals directly with the memory. There is nothing between the chipset and the memory chips on the module as they communicate.
Desktop systems generally use unbuffered RAM.

(Source: Crucial.com)

RAM Cooling
Heatspreaders were originally placed on Rambus. Due to the nature of this RAM type, one IC could heat up more than others, and so heatspreaders were introduced to do exactly as the name would suggest - spread heat evenly along the module.
On DDR, heatspreaders by themselves don't really help. Often the thermal tape used is applied poorly and this makes thermal transfer properties often worse than having no heatspreaders at all, and of course heatspreaders are rendered useless without airflow. Bare chips > Heatspreaders.
Generally speaking, on your average value PC3200 stuff, RAM cooling is nothing more than a modding accessory. If your RAM does not have heatspreaders then you are certainly not at any disadvantage.
Now, if you were to run much over 3v through some Micron -5B-C ICs or some BH-5 for example, or push TCCD/5 to it's limit, then the modules get pretty hot and stability will be compromised. Active cooling really helps here. Placing one or two small fans over the modules is very effective and you only need 20-30CFMs, but even having good case airflow over that area can go some way to helping. If you cannot place your finger directly on an IC (or above where the IC is situated if a spreader is present) indefinitely without having to withdraw your finger due to the heat, your sticks are too hot.

Stuck for ideas on how you could actively cool your RAM?
http://xtremesystems.org/forums/showthread.php?t=47413
http://www.dfi-street.com/forum/showthread.php?t=19832

Other reasons I think manufacturers place heatspreaders over their modules...

If a company is forced to change ICs without prior notice, due to demand/availability/price, there will be uproar from the customers if they buy RAM thinking it's IC X when it's actually changed to IC Y. For example when TCCD became TCC5, the heatspreader was useful because the IC used could not be seen, and as TCC5 would be little known to the average user, they would probably be upset to find it wasn't TCCD. If the heatspreader is there they would be none the wiser and through use of the module would realise there is no difference between TCC5 and TCCD once binned to the correct degree by the manufacturer.
Another obvious reason for a heatspreader is if a company spends a lot of time and money on R&D, then they go ahead and display the IC's to competitors, they will in effect have done the R&D for their rivals by handing their results to them, so then the competitor would be able to create a similar module but would not have incurred the R&D costs.

Well, the above is what the manufacturer intends, but people always find out. Even if the IC is relabelled you can work out what it is from physical or performance characteristics.

Dual Channel RAM
DC as a feature was first introduced to the majority of consumers with the nforce2 chipset. With the original nforce2 concept Nvidia wanted to create an integrated video solution. With integrated graphics DC would help since it could feed off one of the channels. But Nvidia missed the price point for an integrated graphics solution and made another version of the nforce2 chipset minus the integrated video. They then tried to push DC as a performance enhancing feature but we all know that DC on SoA makes no difference greater than perhaps 2-3% in games because the way in which the CPU communicates with the memory controller is too slow to take advantage. With Intel systems DC has a larger impact because of the greater bandwidth available, and it's the same with the A64 because of the IMC (integrated memory controller). Memory companies then took advantage of all the hype surrounding DC, tricking the consumer into thinking they need special "Dual Channel RAM" to run in dual channel mode. But the ability to run DC is down to the memory controller, not the RAM. To run in DC you simply need a memory controller than supports it and to ensure that the individual channels total the same capacity. You do not need the modules to be the exact same brand, revision or timings, although it may help with compatibility you shouldn't normally have a problem mixing modules. Dual channel kits are marketing. The modules contained within are the same model, tested to run together, and often have corresponding batch numbers, though these features are not essential to run in DC. However, since most RAM comes in these packs now anyway, you may as well buy one rather than two individual modules.
Note: Contrary to common belief, even in a DC kit, one stick will always OC differently to the other.

The Relationship Between Cache Size And RAM Capacity
The relationship is..... there is no relationship! No way. Nada. Not a sausage.
So if anybody ever writes something like this:

"your processor has only 512K cache, that means the max optimum amount of memory that you can have in your system is 1GB, So going 2gb you can actually lose frame rates in games"

...They do not know what they are talking about!

Certain RAM Is Optimised For A Certain Board (And Vice Versa) ? Eh ?
Recently I have noticed Asus and Corsair forming an alliance where by apparently Asus' boards are optimised for Corsair RAM and Corsair RAM runs best on Asus boards. Another marketing gimmick, lovely. All major mainboard manufacturers work closely with the bigger DRAM manufacturers anyway. This is why some boards default to a 2T command rate, because the DRAM manufacturer has requested this to ensure better stability / compatibility with their modules. I'm sure Corsair works great on Asus, but I'm also sure it works great on most other boards too !

Testing Your RAM
www.memtest.org

Create a bootable floppy or ISO. If you don't know how to do that I suggest you Google. Then check your boot order so that memtest will run from your chosen source just after POST and before booting into your OS. This is ideal since the risk of corrupting system files when booting into Windows on an unstable FSB/HTT/RAM overclock are rather high. (Simply upping the CPU multiplier to increase clock speed on a default or supported FSB will not lead to corruption). So to avoid this you should always run memtest when pushing the FSB/HTT or playing with your DRAM config.
If trying to determine whether your RAM is faulty or not at stock speeds you should run through the entire test (7 stages) indefinitely. Perhaps leave it running overnight or when at work. You should not see any errors. Even one error is one too many.
If overclocking, I would recommend running a round of memtest each time you increase your HTT. I would suggest looping #5 as you're most likely to encounter errors in this particular test. To loop test 5 press C on your keyboard to get to the configuration page and then 1, 3, 5, enter, 0. Loop test 5 for 10 passes. If you've no errors that's good. If you've under 50 errors you can probably tune your way out of it by adjusting DRAM settings, or switching slots. If you've many many errors, you've probably pushed too far, in which case there's not a lot that can be done to remedy that except obviously buy better components.
Note 1: Passing memtest does not guarantee stability in Windows. So, if you make it to Windows, successfully complete SuperPI 32M if you want to prove that you're stable.
Note 2: When finding your limits, be aware that the above tests do not guarantee 3d stability for heavy gaming.

Speed-Binning
I use this phrase all the time, usually shortened to "binned" or something, and maybe some people wonder what I'm talking about.

Basically, all IC's used in a particular model/revision are the same, yet some IC's are capable of higher speeds than others. Manufacturers test them and throw them into different speed "bins" depending on how fast they run. (Interestingly, the intensity of speed-binning or IC sorting done by the manufacturers is quite often the reason for differences in price between seemingly similar modules.)

Speed-binning and the consumer:

1) Intense binning generally ensures very capable modules which overclockers expect. If you want modules capable of 275MHz or PC4400, you get modules binned to that level. They are pre-tested, if you like, to run at those speeds.

2) If for example Corsair were creating a new PC4000 kit, and they had a no win situation with a batch of IC's -- maybe the IC's only did 240MHz instead of 250MHz -- Corsair would not want to lose out on the IC's they purchased so they then use them in Corsair Value modules instead. The reason being to make a little profit on a lot of IC's is of course better than making no profit on anything at all. Consequently, the unsuspecting Corsair Value buyer finds out his modules run to 240MHz.

The Page File
Running multiple applications simultaneously or playing the latest games with their large textures consumes vast amounts of memory. When you run out of memory, your page file comes into play. The page file is stored, by default, on your boot partition although it can and perhaps should be moved for improved performance to another drive (or at least partition) so that Windows can handle read/write requests quicker. When you run out of physical memory, Windows uses this page file as it were physically memory. Unfortunately it's a lot slower than RAM, so if for example you were playing a game and the game ideally required more RAM, you would notice constant hard drive access due to the page file and therefore your game would stutter. If you ran out of physical RAM whilst in Windows, your desktop would crawl because of the constant HDD access. This is why, up until a point (2GB for most people here) more RAM is beneficial. It reduces or eliminates the page file access which will in turn make for a much smoother and more responsive system. So essentially if you notice your games becoming jerky/stuttery when entering a new area of the map for example in an FPS, take a look at the hard drive access LED on the front of your case. If it's flashing when you notice the stuttering then you could use some more RAM. Typically your page file size, both max and min, should be 1.5x your system RAM. So if you have 2GB RAM, a 3GB/3000MB page file.

Last edited by WildStyle; 12-07-2006 at 06:14.
   
 
Old
  (#2)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default 07-03-2004, 19:13 | posts: 15,302 | Location: UK

Want RAM? (By Royicus)
When the time comes, and it does for all of us, that you want/need some new RAM, how do you decide what to get? There are tons of sticks out there with a lot of manufacturers. Is there a good way to decide which one to get? The answer, of course!

To start off, when you are looking for RAM, the first thing you should do is determine the purpose for this memory. There are basically two choices, running stock speed and overclocking, with the latter being much more exciting .

I Just Want to Run Stock . . .
If you are just looking to run stock, then getting expensive RAM really turns out to be a waste of money. RAM with tighter timings doesnít make a significant difference in system performance, maybe only a few percent. Money would be better spent getting a better processor, more RAM (to a certain extent), or a better video card. In this case, value RAM would be the best choice, both logically and financially. For those of you that fall into this category, you can just stop reading here Go out and get yourself some low priced RAM from a reputable manufacturer. A reputable manufacturer is important for a few reasons, including potential RMAís. However, donít be scared away from the word ďvalueĒ as it is your friend.

I Want to Overclock!!!!
Woa there, now this is where it gets exciting, and Iíll base the rest of this little guide around this. You know you want fast stuff, but which one? Certainly value sticks wonít cut it as value PC3200 often maxes out around 210, maybe 220 if you are lucky. So what do you get?

Picking RAM for Overclocking
First and foremost, you DO NOT get RAM based on brand. Iíll compare that to getting a sports car without knowing the performance numbers. Sure, a sports car is a sports car, but they arenít all nearly the same. The same is true for RAM. OCZ RAM isnít necessarily better than lesser known brands, such as Kingmax or PQI. You have to shop by what the RAM uses (PCB and ICís), not by brand.

DO NOT SHOP BY BRAND!!!

Ok, so now you know that you shouldnít shop for brand, how to you shop? Well, you pick by the PCB and particularly the ICís that the RAM uses. So what PCB and ICís should I be looking for?

You, ICís, PCB, and Overclocking
There are dozens of ICís and a few kinds of PCBís that are used on RAM, but a few in particular are known as good for overclocking.

Winbond BH-5, UTT CH-5
These sticks are the best you can get at the moment. They run high clocks (250+) at tight 2-2-2-x timings. However, there is one catch. You need to be able to supply these modules with around 3.4V or more to do so. 3.4V is a lot, and only a handful of motherboards can supply this kind of voltage. Make sure you can if you want to get these and push them to their max. If your motherboard canít, there is a DDR Booster out there made by OCZ that, if it fits your motherboard, can supply your RAM with more than enough voltage to hit its maximum clock speed. If the DDR Booster doesnít fit, or you just donít want to get it, there is another alternative...

*Edit by Wild* There are both UTT BH and UTT CH available, with the only difference being that BH requires less voltage at lower speeds, where as CH generally requires at least 3v to POST running 2-2-2-x @ 200MHz. In the end though, they should top out at about the same frequency with the same amount of voltage.

*Edit by Wildy*
UTT IC's explained: http://www.legitreviews.com/article.php?aid=199

EDIT: BH-5 (div) vs. TCCD (1:1) : http://www.dfi-street.com/forum/showthread.php?t=40178


*Royicus' BH-5

Samsung TCCD/TCC5
I would recommend these for the vast majority of general overclockers. Most motherboards can supply a RAM voltage of 2.8 Ė 2.9, which is enough to run this RAM at high clocks. However, TCCD RAM does not run with tight timings at high clocks, so clock for clock, the Winbond RAM mentioned above will outperform it. However, since the required voltage is so much lower, this is the RAM that is appropriate for most. TCCD is actually rated for 250MHz @ 3-4-4-8 by default.

*Edit by Wild* You may have noticed TCC5 being used instead of TCCD. TCC5 is essentially the same as TCCD, except that it is only binned for 466MHz by Samsung rather than 500MHz like the TCCD. They are both "binning code names" of the Samsung K4H560838F IC. However, providing the manufacturers binning process is a good one, there will be no significant difference between TCCD and TCC5 to the end user. (Why are Samsung doing this? To lower production costs to compete with UTT - See link about UTT ICs explained somewhere else on this page).


*Wild's TCCD.

Brainpower PCB
I wonít get into the gritty details of this (You can PM WildStyle if you want to know more), but the PCB that the RAM uses is an important factor if you want the most out of your RAM. Currently, Brainpower PCB is the best to run the aforementioned ICís at their maximum. If it is not based on Brainpower, that doesnít mean that it wonít run fast, but it would probably run faster if it were.

*Edit by Wild* Incidentally, BrainPower PCB's are cheaper to build than the JEDEC reference design. BrainPower B6U808/815 PCB's are recognisable by the 8 components along the lower edge of the back side of the DIMM, just above the slot fingers. Also, if no heatspreaders are present and the ends of the module are visible you will see B6U808 or 815 printed.

*Wild.. I'm still editing* Memory PCB Selection Matters: http://www.legitreviews.com/article.php?aid=53
(Unfortunately tested on Intel so TCCD doesn't shine but use this article purely to see the difference a PCB can make. JEDEC reference vs. Modified JEDEC vs. BrainPower).

So What Now?
Well, now you know how to get your RAM, not by brand. You want to get particular ICís on a particular PCB. Once youíve chosen the IC appropriate for you, what brand should you get?

Brand awareness
You may not have heard of Centon so much, right? Most of you probably even consider them generic. What if I said that "under the hood" the Centon modules I had were better than the Corsair XL everyone loves. Both using TCCD. The clocking ability is better than the XL since XL rev1.1 was crap. It would be the same for XL rev1.2 as well though. XL rev1.2 is pretty good, equally as good as the Centon but I know people would still pay more for, and prefer to have the Corsair just because it's Corsair. Makes little sense. That's why you all need to understand that brand really is not top priority. You can save a lot of money by picking RAM that isnít flashy, isnít filled with LEDs, and may not be as well known to all.

*Edit by Wild*
An example of some RAM many may consider boring to look at it is this G.Skill...



But actually, it's bloody good stuff !

RAM Reviews and What You Need to Know
We all read them, and some make our jaw drop with what they achieve. What are they? Insane RAM overclocks on some OCZ, GSkill, Mushkin, or the like. However, does your RAM run 300+ MHz like you read about online? No. Why not?

When RAM manufacturers send samples to online groups to review, they are more often then not hand-picked modules. Out of all the ICís out there, the sticks they send to whomever to review will be in the top fraction of a percent in overclockability. Naturally, the review is very glowing, praising the new RAM about how beautifully it works, which is what the manufacturer wants to see when promoting a new product.

Donít be fooled by these reviews, or some staggering numbers that you see on other forums. The average user wonít be able to obtain these ďspecialĒ modules that the big names in overclocking get. Donít get me wrong, you can end up with a set that does just as well as some of the best results out there, because the manufacturer wonít sort through ALL their ICís for their testers, but your chances of purchasing such a set through pure luck is very slim. You see some of us say this around the forums, but overclocking is the luck of the draw. Itís all about probability. Getting specifically binned ICs rated to run at 250MHz is one way to make the odds lean in your favor, for example.

If you are concerned about how well a particular kit overclocks, Iíd recommend you scour some forums and talk to some knowledgeable users, and see what kind of overclocks they have been seeing. This will ensure that you get a kit of RAM that you will be relatively happy with, rather than feeling hard done by.

Best of luck in getting that magic set. I know I didnít.

Conclusion
So go out there and find the manufacturers that use the IC and PCB that you want. Then, pick the lowest price from a reputable brand. Purchase it. All RAM that Iíve seen comes with a lifetime warranty. If it ever goes bad, send it back to the manufacturer for a new stick. Now, this is where a brand could potentially make a difference. Some companies are better with the RMA process than others, so that could be a factor in picking one brand over another. Generally, companies are good with replacing your RAM if it is faulty, but this is something to think about if for some reason, RAM always breaks on you


So there you have it. The details on how you should go about buying some new RAM. Go out, make a good purchase, and enjoy great performance!!


*Edit by guess who*
I would like to know which IC's my RAM is using.
This list should help if it's a known module where the IC would be of concern: http://www.xtremesystems.org/forums/...ad.php?t=50010

If you want to know a particular ICs characteristics...
http://www.vrforums.com/showthread.php?t=47615
Or talk to me (Wild)

Last edited by WildStyle; 03-30-2006 at 23:58.
   
Old
  (#3)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default 07-04-2004, 14:36 | posts: 15,302 | Location: UK

Stuff I Couldn't Fit In The First Post Because I Ran Out Of Space

DFI NF4 BIOS
Newer BIOS' for these boards will come in 3 versions. You will, for example, notice a 623-1, a 623-3 and a 623-3. (BTW, boards will always ship with the latest official general/unoptimised BIOS i.e. if you bought an nF4 Ultra-D now it'd arrive with the 623-1 BIOS).

-1 = General BIOS for all RAM. Not optimised for any particular type.
-2 = Optimised for BH/UTT BH/UTT CH.
-3 = Optimised for TCCD/Micron.

BTW, the 623 in the example used above refers to the BIOS being released on the 23rd day of the 6th month (of 2005).

The difference between the three BIOS versions is the ROMSIP table used. A ROMSIP table is the memory controllers internal timings and along with the correct alphas can make a big difference to how your RAM plays. Some configs prefer one ROMSIP type, some another, hence the comments below.

As always, the above is not an exact science, it is merely a rule of thumb. I have seen people running for example TCC5 on a "-2" BIOS with great success yet I have seen people trying to run Micron -5B D in a "-2" or "-3" BIOS and the result was a no POST and a while spent sorting it out. The general BIOS will be fine for most types of RAM, it is really only if you are pushing for stability at high clocks that a BIOS designed for your specific mem type may become useful. With regards to slots, Orange is recommended as they generally work better than yellow, but as always, you should really test for yourself to find out which arrangements work best for you. However, if you're lazy, just follow the recommendations.

If for whatever reason you flash to a BIOS that doesn't like your RAM, and you can't POST or the BIOS hangs, you need to power down, clear the CMOS, put a stick of RAM in the top yellow slot and then power up again. The system should now at least POST, meaning you can then flash to a more appropriate BIOS either from within DOS or via WinFlash in Windows.

1GB DDR1 Modules For Overclocking
See: http://www.xtremesystems.org/forums/...ad.php?t=67762

Micron -5B D based modules are pretty much on top right followed by Infineon BE-5/CE-6 and Samsung UCCC modules. The Ballistix sticks using the -5B D IC's are doing so well because of the comprehensive binning. Ever wondered why they're often out of stock? Because it takes Crucial ages to sort the ICs as they're looking for the best and the best only. Unfortunately, Micron won't sell to Taiwan so you'll have to wait and see what companies such as GeIL come up with.

Note: I would not recommend OCing 1GB sticks on anything prior to the Rev. E mem controller. You will be very dissatisfied with the results. Overclocking 1GB modules is more demanding than OCing 512MB modules and unfortunately older IMC's do not seem to cope very well.

Note #2: TCCX does not come in a density sufficient enough to make a 1GB module.

Note #3: Crucial Ballistix PC4000 1GB sticks (Micron -5b D) suffer from sudden death syndrome. I have experienced this myself. I cannot recommend these modules anymore ! This thread is the largest I have seen on the issue: http://www.dfi-street.com/forum/showthread.php?t=29743

Note #4: Infineon based have poor 3d stability ~260MHz & upwards.

Note #5: Get RAM with tight timings and use a divider. A64's are all about CPU MHz and a divider with tight timings is faster than 1:1 with loose timings since you are actually running a divider all the time anyway ... (see below). Think about it, why else would people use BH5? See here: http://www.dfi-street.com/forum/showthread.php?t=40178

Dividers (General)
A divider is useful when you want to overclock your HTT/FSB, yet your memory isn't made for overclocking. When running the 1:1 or 6/6 or 200Mhz divider, you are in synch with the FSB. This means when you increase your FSB, your memory frequency will increase accordingly. If your memory is holding back your overclock, or to find out if it is, you can run a divider. Basically, this slows your RAM frequency down, allowing you to push your FSB higher without being limited. The degree to which performance drops from running a divider varies between platforms. For example on an AXP, running a divider induces a huge performance penalty, and isn't recommended at all. Opposingly, on an A64 running a divider has next to no effect on performance. A64 dividers are complicated, but that's what I'm going to focus on here.

Dividers (A64 specific)
Have you heard that running a divider on an A64 doesn't affect performance much, but want confirmation? Well, here's the situation. Due to the A64's architecture, it's on-die memory controller, and the fact that A64's are not starved of bandwidth, there is no significant advantage to running your memory "in-synch" or "1:1" anymore as opposed to running a divider. There is a small gain to running "1:1", but if you're able to tighten up the timings or run a higher HTT whilst on a divider, this will cancel out that small gain. http://www.dfi-street.com/forum/showthread.php?t=40178
Notice that I have quoted that "1:1." This is because even at the 200 divider (a.k.a. 1:1) the memory frequency is derived from the CPU's clock speed (as explained below), so you can never truly run 1:1 in the sense that you're used to.
Finally, you really needn't splash out on expensive RAM anymore unless you really want to, because now that you know dividers have little impact on performance, you can save your money if you want to overclock by getting cheaper RAM and then spend that saved money on another part of your system.

So, you have your new A64 setup, but you bought value RAM and it won't OC past 220MHz. Well that's ok, because you can run a divider. You should try and keep your memory frequency as close to that 220MHz as possible, you will get better performance. The divider options you see in the BIOS are 200/166/133. Knowing already than 200 is 1:1, you decide to try the 166 divider. That might work, but you are at 300HTT and you don't know what memory frequency will get set when you choose this divider. You are also too lazy to clear the CMOS if it doesn't POST! The text below will help you work out what memory frequency you are running, but TBH, just use the app!

The divider is applied to the CPU clock. Memory frequency has to be a function of the CPU clock since the memory controller is on the CPU.

Imagine you wanna run 300x10 for 3GHz but your RAM can only achieve 250MHz so you choose to run the 166MHz divider in the BIOS.

RAM multiplier: ceil(200HTT * 10x = 2000MHz / 166MHz = 12 (divider - integer))

RAM speed: 300HTT * 10x = 3000MHz / 12 (divider) = 250MHz (RAM freq.)

So I guess the formula would be:

RAM multiplier: ceil(200HTT * CPU multi = CPU clock / RAM ratio = divider)

RAM speed: HTT * CPU multi = CPU clock / divider = RAM frequency.

Note: ceil = divider rounded up to next whole number.

If you are like me and find all the above does your head in, use a program to help you work it all out: http://www.forumdeluxx.de/forum/atta...6&d=1103300738

I have a headache.

What Happens If I Fill All Four Slots On My A64 939 System?
Unfortunately you'll encounter a flaw in the A64's memory controller. Bear in mind that 400MHz @ 1T is ideal.

Newcastle/Clawhammer/Winchester: 333MHz/2T (Single sided/Double sided - No difference.)

Venice/San Diego: 400MHz/1T (If single sided) 400MHz/2T (If double.)

There is a link in this guide somewhere which benches 1T vs. 2T BTW.

Also, and this may be obvious to some, but the mem controller's gonna have a hard time overclocking four modules simultaneously.

It is better to go for 2x1GB than for 4x512MB.



Latencies/Timings
Latency is the time between when a request is made and the request is answered. I.E, if you are in a restaurant for a meal, the latency would be the time between when you ordered your meal to the time you received it. Therefore, in memory terms, it is the total time required before data can be written to or read from the memory. Thus lower is better.

(Source: Hilbert)

Example..

2-2-2-5 1T

From left to right..

CAS-tRCD-tRP-tRAS CMD


tCL/CAS - Column Address Strobe
Undoubtedly, one of the most essential timings is that of the CAS Latency and is also the one most people can actually understand. Since data is often accessed sequentially (same row), the CPU only needs to select the next column in the row to get the next piece of data. In other words, CAS Latency is the delay between the CAS signal and the availability of valid data on the data pins (DQ). Therefore, the latency between column accesses (CAS), plays an important role in the performance of the memory. The lower the latency, the better the performance. However, the memory modules must be capable of supporting low latency settings.

tRCD - RAS to CAS delay
There is a delay from when a row is activated to when the cell (or column) is activated via the CAS signal and data can be written to or read from a memory cell. This delay is called tRCD. When memory is accessed sequentially, the row is already active and tRCD will not have much impact. However, if memory is not accessed in a linear fashion, the current active row must be deactivated and then a new row selected/activated. It is this example where low tRCD's can improve performance. However, like any other memory timing, putting this too low for the module can result in instability.

tRP - RAS precharge
tRP is the time required to terminate one one Row access and begin the next row access. Another way to look at this it that tRP is the delay required between deactivating the current row and selecting the next row. Therefore, in conjunction with tRCD, the time required (or clock cycles required) to switch banks (or rows) and select the next cell for either reading, writing or refreshing is a combination of tRP and tRCD.

tRAS - Active to precharge delay
Memory architecture is like a spreadsheet with row upon row and column upon column with each row being 1 bank. In order for the CPU to access memory, it must first determine which Row or Bank in the memory that is to be accessed and activate that row via the RAS signal. Once activated, the row can be accessed over and over until the data is exhausted. This is why tRAS has little effect on overall system performance but could impact system stability if set incorrectly.
tRAS is not an exact science, lower is not always better. There was a theory that the fastest tRAS value to use was decided by CAS + tRCD + 2. e.g. if your timings were 2-2-2, the ideal tRAS value would be 6, but I disagree with this. You need to work out for yourself which performs best because it varies from chipset to chipset. The difference will be extremely minimal but if you're going to manually define the tRAS at all you may as well do it properly. You can use SiSoft SANDRA memory bandwidth benchmark to compare. Having the tRAS too tight can also limit your RAM's max frequency, so keep that in mind as well.

Command Rate (CMD)
The Command Rate is the time needed between the chip select signal and the when commands can be issued to the RAM module IC. Typically, these are either 1 clock or 2.
1T is faster, 2T can often help with stability and will be the default in most BIOS as per DRAM manufacturers request.
*CPC (Command Per Clock) : CPC Enabled = 1T, CPC Disabled = 2T.


A couple of quick notes...
tRCD and tRP have the biggest impact on performance. CAS does not. I believe this is mainly because altering the CAS value you are all familiar with adjusts only the read speed. To adjust the CAS write speed also you need to alter the tCWL.

CAS 1.5
CAS 1.5 does not exist. I don't know what gets set, but it's not 1.5! Look below.. see the option for CAS1.5 in the A64's memory controller? Nooo!

000b = reserved
001b = CL=2
010b = CL=3
011b = reserved
100b = reserved
101b = CL=2.5
110b = reserved
111b = reserved

It is a name, that's all, a marketing gimmick. It looks cool, but there is no performance gain at all. In fact, it's often considered slower and this is why you may experience greater stability @ "1.5" than 2. It's similar to tRAS = 00, it doesn't exist because it's reserved but I know for a fact it can add stability at very high TCCX speeds.
You cannot really bench CAS"1.5" vs CAS2, though, since the margin of error is so small.

Don't get your hopes up
When I say "biggest impact on performance" I don't mean you will see major FPS boosts in games. The difference will really be noticable in benchmarks only. A lot of people here will tell you not to even bother with adjusting latencies given how little they effect system performance but think of all the small tweaks you can do to your system to improve performance, and think what they add up too - a nice boost. And to the tweaker/bencher, latencies make a big difference.

To demonstrate my point....

How memory timings affect A64 performance: http://www.xtremesystems.org/forums/...ad.php?t=48634

2-2-2-5 vs. 3-4-4-8: http://www.gamepc.com/labs/view_cont...s3200xl&page=1

2-2-2-5 vs. 2.5-4-4-8 & 1T vs. 2T on A64: http://techreport.com/etc/2005q4/mem...y/index.x?pg=1

Does RAM latency matter? http://www.extremetech.com/article2/...1637762,00.asp

TCCD/DFI nF4
I was gonna post this waaaaay back, since all the RAM I've properly owned/tested in the last year has been TCCD. I just figured there wasn't enough people here into this stuff. But hey, it's my thread, and it might help one or two of you out.

So.. TCCD on a DFI nF4... Firstly you need a new BIOS as opposed to the BIOS your board ships with (-1). It's not that it's bad it's just that you can do better. Any "-3" BIOS (623 onwards) used in conjunction with the orange slots should yield nice results or if you fancy trying out the newer 704-2BT/BTA BIOS' (BTA is improved BT - more relaxed default settings and some options renamed) modded by Tony from OCZ I would advise you use the yellow slots. Personally I use 623-3 on orange.

You can find BIOS' here: http://www.dfi-bios.tk/ and/or http://www.ocforums.com/showthread.php?t=405670

Start @ 2.7v and increase only if all else fails. More often than not a higher amount of vdimm will only give more errors.

On to the timings, firstly you need to read this: http://www.dfi-street.com/forum/showthread.php?t=11397 to familarise yourself with the settings you will be changing. You should be playing with these to further your memory frequency or simply to increase your bandwidth. The timings shown in the guideline below should be good for 300MHz plus provided your RAM is actually capable enough. They are pretty loose and can certainly be tightened up, but this is just to guide you. If you have a Winchester then chances are you will have a weak IMC. (Unfortunately if that's the case you won't know if it's the RAM or IMC limiting you until you swap one of the parts out). Newer cores/revs shouldn't have a problem. This is targetted at 2x512 users since that's what I use. I would enter these @ 200MHz and work your way up in 5/10Mhz stages, memtesting by looping #5 for 10 passes at each increase, as explained earlier in this RAM guide. Good airflow over the modules is a must at high frequencies.

MHz - 200 (1:1)
CPC - Enable (This is 1T. Disabled is 2T)
tCL - 2.5
tRCD - 4
tRAS - 7
tRP - 4
tRC - 12
tRFC - 14
tRRD - 3
tWR - 3
tWTR - 2
tRTW - 2
tREF - 4708 (166MHz 1.95us in A64 tweaker)
tWCL - 1
Bank Interleave - Enabled (When using two double sided modules this is a free performance boost similar to running 1T over 2T).
Skew Control - Increase
Skew Value - 255 (Has little effect on stability & higher = faster).
Drive Strength - 7 (TCCD/5 need weak drive strength. 1,3,5,7 = weak, 2,4,6,8 are normal with 8 being strongest).
Data Drive Strength - Auto
Max Async Latency - 8
Response Time - Normal (Was added in 509 BIOS. If you don't see it your BIOS is much too old).
Read Preamble - 5
IdleCycle - 256
Dynamic Counter - Enable
R/W Queue Bypass - 16x
Bypass Max - 7x
32 Byte Granularity - Disable (4 Burst)

And finally someone was asking me what DRAM response time was because they couldn't find a definition anywhere....

DRAM response time affects the Max Async Latency and Read Preamble settings. If those two settings are on Auto then the DRAM response time sets them based on your "Normal, fast, faster" selection. If you set the MAL and RP manually then the DRAM response time has no effect.

Last edited by WildStyle; 12-23-2006 at 22:13.
   
Old
  (#4)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default 07-04-2004, 18:12 | posts: 15,302 | Location: UK

Micron D9/DDR2
D9 is simply Micron's code for DDR2, so by itself the term doesn't mean anything. However we'll use it as it's a well known term by now.

If you're looking to upgrade to a platform that requires DDR2, you have probably heard about the Micron D9 IC's. They can run to high frequencies and can run tight timings.

There are a huge amount of different steppings/platings/types and speed grades in the D9 family, but let's focus on two particular types, the B6 and the BT.

The BT is the original "Fatbody" D9 which are no longer being produced. The Fatbody's clock better with low timings, D9GMH/D9GKX etc clock better with higher timings.



The B6 is the next best thing and fortunately these are still in production and readily available. So right now if you bought some sticks they are probably going to be D9GMH or D9GKX (the former being much more probable) which are B6. GKX is said to be better and comes binned from Micron at 400MHz as opposed to GMH which is binned to 333, both at 5-5-5. This is irrelevant really though as they'll be binned by the manufacturers anyway, and generally the better quality binning the higher price but the better the module.

Binning or sorting basically means the manufacturer sorts through all their IC's and divides them up into different "bins" based on their ability. Some IC's, even though they're from the revision/model, will perform better than others. If the manufacturer had a high-end model then the top say 5% of ICs would go into this model, and the price would be increased accordingly!



(Images aren't to scale but you can still see the obvious shape difference).

For info on how GMH scales, have a look here: http://www.overclockersonline.net/in...num=732&pnum=5

The best way to find out what sticks use which IC's is to take a look at this list: http://ramlist.ath.cx/ddr2/ (New D9 in green, old D9 in red). It's a thorough list, but not definitive, and I'd say it's always best to check out some of the better known overclocking forums to check that those ICs are still being used, because if you order online you won't know until they arrive and you either take the heatspreaders off to have a look or see how they behave in your system. If you notice some models in the list have a "Handpicked" next to them, that simply means the ICs were sorted through and the best ones they had went into that module. You can also find details on most DDR2 releases here: http://www.xtremesystems.org/forums/...ad.php?t=94017

Some models aren't featured on the above link, so if this is the case with any RAM you're looking at you could either leave it and find something where you know the IC or you could dig around and see what you can find out. As an example of a module that's not on the list (D9 in fact), and also an example of switching ICs, I know John's Corsair 6400C4 are GMH since he took off the heatspreaders and had a look but this model also use Promos ICs as well, dependant on revision. So it's not all cut and dry. Corsair are a major player and it's not practical for them to place the greatest ICs on all their modules. Besides, if they did do this, they would be unable to meet the demand for their PC6400 range, so there is some reasoning behind it. Whilst the Promos used in place of the D9s aren't terrible, if you were expecting D9s and got Promos I imagine you'd be a tad annoyed. Thus it's always best to check. If you can't find any info then I might be able to help.

Note for those looking towards Value RAM hoping it might contain decent D9 (not all D9 delivers the wow factor, you need to look for specific) basically Value RAM is built with whatever ICs are cheapest at the time and this changes on a regular basis. Probably not worth the risk.

Let's touch on voltages and cooling a sec. RAM prices are extremely inflated right now because the DRAM manufacturers (Micron, Samsung, Hynix, Infineon etc) jacked the prices up and so the companies that build the modules (your OCZs, Corsairs etc) are having to adjust their prices accordingly. So with that in mind once you have a set of this stuff you should probably try and look after it as well as you can. I would personally recommend you stick to under 2.4-2.5v when overclocking the D9s for 24/7 use. Active cooling is highly recommended. For benching, yes you can take them upto 3v if you have the ability but don't think about running those sort of voltages 24/7 as the RAM will die. Now theoretically the Fatbodys can handle more voltage since they're built on a larger manufacturing process (90nm vs. 80nm) but in reality I don't know if that holds true, I wouldn't want to risk it as if you can get hold of Fatbodys you want to treat them like gold dust pretty much.

When I said active cooling recommended, I meant it. Throughout my time here I've noticed everyone seems to avoid using active cooling on RAM because they probably don't understand how much it helps no matter how much I say it. They also probably think it'd look a bit crap to have to rig a fan up with elastic bands, cable ties, Zalman fan brackets, etc. *shrug* Well now you don't need to because Corsair and OCZ both offer devices for the job...

http://www.overclockers.co.uk/showpr...odid=MY-123-CS

http://www.overclockers.co.uk/showpr...odid=MY-083-OC

There we go, easy. If however you're even the smallest bit resourceful you can rig something up pretty easily and avoid paying 20 quid for a pre-made RAM cooler. Right now I personally have an 80mm YSTech blowing over my RAM which you can see below:



The piece of plastic holding the fan is actually one of those 5.25" drive rail things you get with some cases. Works fine.

DDR2 Speed Grades/Ratings
PC2-3200 / DDR2-400 / 400MHz
PC2-4300 / DDR2-533 / 533MHz
PC2-5400 / DDR2-667 / 667MHz
PC2-6400 / DDR2-800 / 800MHz
PC2-8000 / DDR2-1000 / 1000MHz

Core2Duo Allendale/Conroe = 1066 / 4 = 266MHz FSB. 266x2 = 533Mhz = you need PC4300 for above CPU, higher rating or known good overclocking ICs if you want headroom to overclock.

Last edited by WildStyle; 12-09-2006 at 23:18.
   
 
Old
  (#5)
xankazo
Ancient Guru
 
xankazo's Avatar
 
Videocard: EVGA GTX 1060 SC (6GB)
Processor: Intel Core i7 4770
Mainboard: Asus Maximus VI Hero
Memory: Kingston HyperX FURY 16GB
Soundcard: SupremeFX / Z-5500
PSU: Seasonic G Series 550W
Default 07-05-2004, 02:42 | posts: 2,803 | Location: Santo Domingo

sticky all the way!
   
Old
  (#6)
Heat-40
Master Guru
 
Heat-40's Avatar
 
Videocard: MSI 980 ti
Processor: I7 4790K
Mainboard: ASRock Z97 Anniversary
Memory: DDR3 Corsair Vengence 16g
Soundcard: Razer Kraken 7.1 chroma
PSU: Corsair Rmi750
Default 07-09-2004, 23:42 | posts: 305 | Location: Scotland

what would be the best for my memory
just now its 2.5/3/3/7
i had it at 2/3/3/7 but i have no if that is good or bad also i see u say 11 for nf2 boards

memory iss 333 fsb @ 2.5v

Last edited by Heat-40; 07-09-2004 at 23:47.
   
Old
  (#7)
Thor
Master Guru
 
Thor's Avatar
 
Videocard: x850XT PE + V5 5500
Processor: P4 3.4 oc 3.6
Mainboard: P4C800E-D
Memory: 2G Corsair pc3700
Soundcard: SB Audigy2 ZS
PSU: TruePower 550
Thumbs up 07-10-2004, 00:00 | posts: 531 | Location: The Clouds with Thunder and Lightni

Great Stuff WildStyle

Here is a link I used to help me understand things also:

http://corsairmemory.com/memory_basi...707/index.html

   
Old
  (#8)
Shex
Maha Guru
 
Shex's Avatar
 
Videocard: GALAX GeForce GTX 970 HOF
Processor: Intel Core i5
Mainboard: Gigabyte z77
Memory: G.Skill Ripjaws - (2x4GB)
Soundcard: Logitech 2.1 | onboard
PSU: OCZ - ZX1000W
Default 07-10-2004, 00:02 | posts: 1,606 | Location: uk

I've got one question to clear up... how "low" can you go with ram timings ??
   
Old
  (#9)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default 07-10-2004, 00:16 | posts: 15,302 | Location: UK

Quote:
Originally posted by Heat-40
what would be the best for my memory
just now its 2.5/3/3/7
i had it at 2/3/3/7 but i have no if that is good or bad also i see u say 11 for nf2 boards

memory iss 333 fsb @ 2.5v
What brand is your RAM, and more importantly, what chips is it using? If it's not rated for Cas2 and it's not winbond, then it's unlikely to do Cas2. It's like my twinmos.... won't do anything better than what it was designed for, timing wise, no matter what vdimm.
   
Old
  (#10)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default 07-10-2004, 00:17 | posts: 15,302 | Location: UK

Quote:
Originally posted by Thor
Great Stuff WildStyle

Here is a link I used to help me understand things also:

http://corsairmemory.com/memory_basi...707/index.html

Yeah, that Corsair site is very good.
   
Old
  (#11)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default 07-10-2004, 00:19 | posts: 15,302 | Location: UK

Quote:
Originally posted by Shex
I've got one question to clear up... how "low" can you go with ram timings ??
As in the fastest timings? 2-2-2-5 are the lowest. But only BH5's and the new Samsung stuff that Corsair etc are now using can do that.
   
Old
  (#12)
Heat-40
Master Guru
 
Heat-40's Avatar
 
Videocard: MSI 980 ti
Processor: I7 4790K
Mainboard: ASRock Z97 Anniversary
Memory: DDR3 Corsair Vengence 16g
Soundcard: Razer Kraken 7.1 chroma
PSU: Corsair Rmi750
Default 07-10-2004, 00:26 | posts: 305 | Location: Scotland

memory is made by MOSEL i had it at 2.3.3.7 and did not get any probs in windows
this is the info from cpuz
Memory type DDR-SDRAM
Manufacturer (ID) MOSEL (4000000000000000)
Size 512 MBytes
Max bandwidth PC2700 (166 MHz)
Part number

Attributes
Number of banks 2
Data width 64 bits
Correction None
Registered no
Buffered no

Timings table
Frequency (MHz) 133 166
CAS# 2.0 2.5
RAS# to CAS# delay 3 3
RAS# Precharge 3 3
TRAS# 6 7

Last edited by Heat-40; 07-10-2004 at 00:30.
   
Old
  (#13)
Shex
Maha Guru
 
Shex's Avatar
 
Videocard: GALAX GeForce GTX 970 HOF
Processor: Intel Core i5
Mainboard: Gigabyte z77
Memory: G.Skill Ripjaws - (2x4GB)
Soundcard: Logitech 2.1 | onboard
PSU: OCZ - ZX1000W
Default 07-10-2004, 00:33 | posts: 1,606 | Location: uk

So you can't go even lower ??

i was just wondering..
   
Old
  (#14)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default 07-10-2004, 00:37 | posts: 15,302 | Location: UK

Quote:
Originally posted by Heat-40
memory is made by MOSEL i had it at 2.3.3.7 and did not get any probs in windows
this is the info from cpuz
Memory type DDR-SDRAM
Manufacturer (ID) MOSEL (4000000000000000)
Size 512 MBytes
Max bandwidth PC2700 (166 MHz)
Part number

Attributes
Number of banks 2
Data width 64 bits
Correction None
Registered no
Buffered no

Timings table
Frequency (MHz) 133 166
CAS# 2.0 2.5
RAS# to CAS# delay 3 3
RAS# Precharge 3 3
TRAS# 6 7
Well, if it booted at Cas2 at 166MHz then that's a good start. First off, try raising the vdimm to 2.8v (help with stability - safe voltage too) and then run memtest ( http://www.memtest.org/ (bootable floppy)) for like an hour, and see if you have any errors at Cas2. You may not have had any probs in windows, but did you do anything intense to test it properly?
   
Old
  (#15)
Heat-40
Master Guru
 
Heat-40's Avatar
 
Videocard: MSI 980 ti
Processor: I7 4790K
Mainboard: ASRock Z97 Anniversary
Memory: DDR3 Corsair Vengence 16g
Soundcard: Razer Kraken 7.1 chroma
PSU: Corsair Rmi750
Default 07-10-2004, 00:38 | posts: 305 | Location: Scotland

no idea i dont know what i should try
   
Old
  (#16)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default 07-10-2004, 00:38 | posts: 15,302 | Location: UK

Quote:
Originally posted by Shex
So you can't go even lower ??

i was just wondering..
Sure. Well they are the fastest timings available. 1-1-1-3 is a no go I'm afraid.
   
Old
  (#17)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default 07-10-2004, 00:40 | posts: 15,302 | Location: UK

Quote:
Originally posted by Heat-40
no idea i dont know what i should try
Anything.... memtest, the 'blend' test on prime95, your favourite game...

There's a good chance it'll just lock up when running a 3d app such as a game if it's not stable. If you're running prime or memtest then you'll jsut get errors.
   
Old
  (#18)
Thor
Master Guru
 
Thor's Avatar
 
Videocard: x850XT PE + V5 5500
Processor: P4 3.4 oc 3.6
Mainboard: P4C800E-D
Memory: 2G Corsair pc3700
Soundcard: SB Audigy2 ZS
PSU: TruePower 550
Cool 07-10-2004, 02:07 | posts: 531 | Location: The Clouds with Thunder and Lightni

To WildStyle - The Last Don,

I bow humbly to your honor.

I have the Corsair site I bookmarked from suggestion of great ones like yourself.

It helped as do all things I find here at this Forum of Guru3D.

   
Old
  (#19)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default 07-10-2004, 13:23 | posts: 15,302 | Location: UK

Hehe, I'm overwhelmed.
   
Old
  (#20)
xankazo
Ancient Guru
 
xankazo's Avatar
 
Videocard: EVGA GTX 1060 SC (6GB)
Processor: Intel Core i7 4770
Mainboard: Asus Maximus VI Hero
Memory: Kingston HyperX FURY 16GB
Soundcard: SupremeFX / Z-5500
PSU: Seasonic G Series 550W
Default 07-10-2004, 21:31 | posts: 2,803 | Location: Santo Domingo

so, raising my vdimm to something like 2.8v could allow my cas to go lower without compromising stability?

right now it's set up like 2.5-3-3-7. It was at 2-3-3-5 but I had some stability issues, however I was too lazy to check if it was my cas timings or other aspects of my overclocked system.

I have two Corsair Value Select sticks hooked up on my board running dual channel.
   
Old
  (#21)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default 07-10-2004, 21:41 | posts: 15,302 | Location: UK

Hmm, well Corsair Value is okay, but I'm not sure if it'd run Cas2... you'll just have to try. But yeah, if it will run it, raising the vdimm would help.
Knowing that it boots at Cas2 is a good sign however, my Twinmos won't even do that.
   
Old
  (#22)
Heat-40
Master Guru
 
Heat-40's Avatar
 
Videocard: MSI 980 ti
Processor: I7 4790K
Mainboard: ASRock Z97 Anniversary
Memory: DDR3 Corsair Vengence 16g
Soundcard: Razer Kraken 7.1 chroma
PSU: Corsair Rmi750
Default 07-12-2004, 15:52 | posts: 305 | Location: Scotland

i ran mem test for 50 mins with my memory at 2.3.3.6@2.6v and go no errors is that ok or should i do anything else
   
Old
  (#23)
Mark
Maha Guru
 
Mark's Avatar
 
Videocard: Sapphire HD4870
Processor: Intel QX9650@4.1Ghz
Mainboard: Asus Rampage Extreme
Memory: 2x2GB G-Skill PI DDR3
Soundcard: Creative X-Fi Xtreme Musi
PSU: Enermax Galaxy 1000W
Default 07-12-2004, 16:39 | posts: 1,617 | Location: Portsmouth UK

Quote:
Originally posted by Thor
Great Stuff WildStyle

Here is a link I used to help me understand things also:

http://corsairmemory.com/memory_basi...707/index.html

wicked presentation. really useful

So was your thread Wildstyle. good job mate
   
Old
  (#24)
WildStyle
Ancient Guru
 
WildStyle's Avatar
 
Videocard: MSI GTX 970 Gaming 4G
Processor: Intel Core i7 2600
Mainboard: MSI H77MA-G43
Memory: 16GB Samsung PC3-12800
Soundcard: Onboard
PSU: Corsair CX600M
Default 07-12-2004, 18:54 | posts: 15,302 | Location: UK

Quote:
Originally posted by Heat-40
i ran mem test for 50 mins with my memory at 2.3.3.6@2.6v and go no errors is that ok or should i do anything else
In that case, it should be okay then. Enjoy.
   
Old
  (#25)
Mark
Maha Guru
 
Mark's Avatar
 
Videocard: Sapphire HD4870
Processor: Intel QX9650@4.1Ghz
Mainboard: Asus Rampage Extreme
Memory: 2x2GB G-Skill PI DDR3
Soundcard: Creative X-Fi Xtreme Musi
PSU: Enermax Galaxy 1000W
Default 07-12-2004, 19:12 | posts: 1,617 | Location: Portsmouth UK

i have infinion ram, does any1 know how well they oc?
   
Closed Thread

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump



Powered by vBulletin®
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
vBulletin Skin developed by: vBStyles.com
Copyright (c) 2017, All Rights Reserved. The Guru of 3D, the Hardware Guru, and 3D Guru are trademarks owned by Hilbert Hagedoorn.