AMD Phenom II X4 920 and 940 review [Guru3D]

Discussion in 'Frontpage news' started by Guru3D News, Jan 8, 2009.

  1. Stormyandcold

    Stormyandcold Ancient Guru

    Messages:
    5,872
    Likes Received:
    446
    GPU:
    RTX3080ti Founders
    The power on offer is monstrous from either P2 or I7 for most users, although; the cost difference is undeniable. In £ it's still around 250-300 difference.
     
  2. Dustpuppy

    Dustpuppy Ancient Guru

    Messages:
    4,146
    Likes Received:
    0
    GPU:
    integrated - fffffffuuuuu
    Well this is pretty much a blow below the belt to the i5 line of processors. Amd is trying to do to intel what they did via ati to nvidia.
     
  3. Modzilla

    Modzilla New Member

    Messages:
    4
    Likes Received:
    0
    GPU:
    GTX 260 Core 216
    From page 11 of the review: "Also compare a little to the Q6600 processor as that pretty much was "THE" quad core processor to get a year ago."

    I have 1 simple question regarding the test method. Was that Q6600 tested Clock for Clock against the new Phenom II speeds or is the benchmark figure given based on a stock clock speed of 2.4ghz?

    Im not just talking about this article, but im talking about all of the other benches on Guru3d where a Q6600 was added as a "comparison" cpu. There is never any mention that the Q6600 was clocked @ 2.8ghz and 3.0ghz respectively for these tests, and IMO that is one really flawed comparison if those #'s are in fact 2.4ghz bench scores.

    Don't try to argue the fact that they are "just trying to show the difference between stock vs stock speeds", because if it isn't clock for clock then it ain't worth the digital graph its printed on. (Besides we all know the Q6600 can OC to 4ghz and beyond)

    On the other hand if it is clock for clock 2.8ghz vs 2.8ghz then I retract everything i've said, but I have the sneaking suspicion that they are using the Q6600 stock bench values.

    Anyone who works @ Guru3d care to emphasize on this?
     
  4. Aura89

    Aura89 Ancient Guru

    Messages:
    8,413
    Likes Received:
    1,483
    GPU:
    -
    uh...what? seriously. there's another one of you crazy people in here?

    So, by your standards, a 9800 GTX+ can't be compared with an 8800 GT because they arn't the same clocks, however, if you make the 8800 GT the same clocks as a 9800 GTX+, then you can see which one is better?

    That's exactly what you just said here, "you can't compare a Q6600 to a Phenom II 920, unless they are clock for clock comparisons", if you wanted that, then what you really would need is "Intel Core 2 Quad @2.8Ghz vs Phenom II 920", because then you wouldn't need to know what CPU it was that Intel had, just had to know it was an Intel Core 2 Duo @ 2.8Ghz

    Why would someone say hey i'm going to compare a Q6600 to a Phenom II 920, to see which one is better for your money, then say, but wait, i'm going to make the Q6600 at Mhz that it's not? What anyone is looking for by these tests are "Hey, i can either get the Q6600, or the Phenom II 920, which one would be better for what i pay for and total performance"

    What you want is a "Core 2 Quad @ 2.8Ghz vs Phenom II 920" comparison, not a "Q6600 vs Phenom II 920" comparison, as you're not making any sense saying it's "not fair" to run it at stock speeds...because, again, that'd be like saying it's not fair to compare a 8800 GT to a 9800 GTX+ unless they are at the same clock speeds
     

  5. Modzilla

    Modzilla New Member

    Messages:
    4
    Likes Received:
    0
    GPU:
    GTX 260 Core 216
    :stewpid:
    Aura89 I said I wanted a clock for clock comparison. You go off on a tangent and start talking about GPU's.....

    Everyone else Check this (its a quick read)
    http://forums.amd.com/game/messageview.cfm?catid=259&threadid=106233&enterthread=y

    As per Firestrider: I was refering to the Q6600 beats the Phenom II 940 given the same clock speed, and this is what the original poster wanted to know.

    Here is the results from that series of benchmarks:
    http://picasaweb.google.com/KT...e#5287403595643346610

    On Average:
    Deneb is 5.2% faster than Agena clock for clock
    Kentsfield is 2.64% faster than Deneb clock for clock
    Yorkfield is 5.6% faster than Deneb clock for clock
    Bloomsfield is 21.05% faster than Deneb clock for clock.

    Thats all I needed to know. AMD & Aura89 FAIL
     
  6. Aura89

    Aura89 Ancient Guru

    Messages:
    8,413
    Likes Received:
    1,483
    GPU:
    -
    :stewpid::bang::puke2::banana::stewpid: oh did i mention, "ModZilla" fails? thanks :) and stop asking for the equivalent of, again, "it's pointless to test an 8800 GT vs a 9800 GTX+ unless they are at the same clock speeds" blah blah blah blah blah, :spam:

    :flip2: durr

    [​IMG]
     
    Last edited: Jan 16, 2009
  7. Cybermancer

    Cybermancer Don Quixote

    Messages:
    13,795
    Likes Received:
    0
    GPU:
    BFG GTX260OC (192 SP)
    Maybe Modzilla has a different opinion/point of view than you Aura, but there's no reason to respond like this:

     
  8. DES_MX

    DES_MX Guest

    Messages:
    769
    Likes Received:
    1
    GPU:
    GTX480 850/4400 1.075v
    Dammit Aura89, if you're too broke to afford something better and u're protecting and trying to justify your hard earned cash invested in an AMD platform, so be it. But don't start posting these ugly and insulting pictures all over the forums.

    Sure you can't compare a 9800GTX with a 9400GS, because they have different number of ROPs and Shaders, use very different memory types and speed.

    However, it's far more interesting and fair to compare a recent generation quad core cpu to another one, and then put them on equal clocks to see which architecture is more efficient. If you're too **** to get this, please just shush.
     
  9. DES_MX

    DES_MX Guest

    Messages:
    769
    Likes Received:
    1
    GPU:
    GTX480 850/4400 1.075v
    P.S. To each his own, and while some people on this thread try to get the best value for their money (PhenomII/Q9550/Q6600), others aim at top-notch performance and complete satisfaction. I have pursued best value for the buck for very long now, but since I've got some extra cash I decided to put a good use to it, and for once get a system I really like. In this sense - core i7 offers the best performance today. So really, stop fighting and make your own decisions. The information is here thank god, so read, reread, contemplate and make a decision. Hopefully it will be the right one.
     
  10. Aura89

    Aura89 Ancient Guru

    Messages:
    8,413
    Likes Received:
    1,483
    GPU:
    -
    eh? poor? whom said anything about money? rofl all i'm saying is don't come in here saying it's pointless to compare a Q6600 to a Phenom 920 or 940 at sock speeds for both of them, when that makes absolutely no sense, there's a very valid point: Does Phenom 920 or 940 beat the Q6600, however, if you want to go ahead and ask "Can you overclock the Q6660 to a Phenom 920 or 940 speed so we can see if Core 2 Quads beat the new Phenom II's clock for clock" that would have made sense, but he didn't, he said, it's pointless to compare a Q6600 to a Phenom 920 or 940, and rofl, that's just stupid
     
    Last edited: Jan 16, 2009

  11. Aura89

    Aura89 Ancient Guru

    Messages:
    8,413
    Likes Received:
    1,483
    GPU:
    -
    k, sorry
     
  12. Iarwain

    Iarwain Banned

    Messages:
    3,047
    Likes Received:
    0
    GPU:
    4890 990//1120
    All of this is just an argument between you geniuses saying "You should review in a style that's most useful for my type of buyer!" "No, you should review in such a way that is most useful for me!"

    Some of you want to compare overclocked results, some of you don't, some of you want to compare regardless of price, others say price doesn't matter, you should review based on the "performance crown," some say that the low-end i7s overclock to the same as the high end ones, so why bother comparing the phenoms to anything but the high end ones.

    At the end of the day, half of you are going to end up unhappy. Why? Because not everyone overclocks, not everyone has the same goal for a processor, not everyone wants to pay the extra dollars for the top of the line, sometimes people just want to see how something stacks up. It doesn't have to be the BEST, just to see it's performance realitive to known quantities, so that individuals can make their own decision about what they should buy for the setup they have and the parts they need.

    And this argument about using the Q6600 being inappropriate is asinine. It wasn't put in there to prove AMD's domination over intel, it was put in there because the Q6600 is a known quantity for most people. People can think "This is how that chip performs, and if the P2 is x amount better, then it's -this- much better than a chip I am familiar with.
     
    Last edited: Jan 16, 2009
  13. DES_MX

    DES_MX Guest

    Messages:
    769
    Likes Received:
    1
    GPU:
    GTX480 850/4400 1.075v
    This is the post where you imply that you haven't had the money. I think you have a personality disorder really proving to me that you never said that.
     
  14. Modzilla

    Modzilla New Member

    Messages:
    4
    Likes Received:
    0
    GPU:
    GTX 260 Core 216
    I agree with you about the fact that Guru3d cannot please everyone. But you have to see the logic in a clock for clock comparison..... Most people won't bother to OC the least bit and just want to know which CPU will give them the best performance/price out of the box. Whereas other enthusiasts want and need to know information like which CPU performs better at a sustainable overclock. You're right Guru3d can't make everyone happy.

    What they can do is at least post the stock clock the rest of the CPU's were tested at. Instead Guru3d just places a stock Q6600 bar and says "This was the go to CPU a year ago".....that is a blatant lie.

    I am not an intel or amd fanboy. I've purchased intel then amd then intel (wish I had gotten amd instead of P4) then intel again. Im happy I purchased my Q6600 and I have no regrets even after the release of Bloomfield.....Kentsfield and Yorkfield are still holding very strong against Bloomfield.....but that will change 6 months from now when newer GPU's are released, since every game is still GPU limited.

    The simple fact remains Deneb was released to compete against i7 and it has failed. In fact it can barely hold its own against Kentfield and Yorkfield let alone i7. Just another graph to prove my point http://www.erodov.com/forums/perfor...a-vs-kentsfield-vs-yorkfield-vs-i7/15525.html

    Sorry AMD....maybe next time.
     
  15. Modzilla

    Modzilla New Member

    Messages:
    4
    Likes Received:
    0
    GPU:
    GTX 260 Core 216
    Thanks Cybermancer.........I don't like being called crazy when all I simply asked for is if Guru3d did clock for clock comparisons....I know now that they do not.

    I think i've made my point about clock for clock comparisons clear and I think i'll just ignore whatever the heck that was that Aura posted......that was even more childish then I thought it would be.....lol
     

  16. chaotic1

    chaotic1 Ancient Guru

    Messages:
    2,841
    Likes Received:
    0
    GPU:
    PowercolorHD6970 2GB dead

    since when ? , since they announced the phenom II i have never read that
     
  17. Denial

    Denial Ancient Guru

    Messages:
    14,206
    Likes Received:
    4,118
    GPU:
    EVGA RTX 3080
    Deneb was released to be an easy upgrade for AM2 users. I worked on Shanghai when I interned with AMD - most of the R&D money was actually spent on making these guys just be able to drop in to your current system/server without issue. I mean - maybe Deneb was also focusing on performance, but it was mainly just a drop in upgrade.

    But yeah, AMD is pretty far behind in terms of having the fastest consumer processor.
     
  18. MileHighMilitia

    MileHighMilitia Master Guru

    Messages:
    601
    Likes Received:
    0
    GPU:
    HIS HD4850 + TT DuOrb 700/1135
    Dont be ridiculous, Phenom II competes with similarly price Core 2 Quads, and it does so well. I will not go into depth about this AGAIN, but a Core i7 system is WAY more expensive, thus putting it and the Phenom II in different price ranges.
     
  19. DES_MX

    DES_MX Guest

    Messages:
    769
    Likes Received:
    1
    GPU:
    GTX480 850/4400 1.075v
    You imply this in non-comparable conditions. Namely:
    1) You imply that upgrading to Phenom II is per se from an existing platform, so you don't have to buy a motherboard and memory. This is just not true. Not everyone upgrades incrementally. A lot of people just wait till their computer has served it's days and get a whole new system. They might hold on to their sound card (e.g. X-Fi is still good); their DVD-RW and their HDD's if they are fast. The rest is exchanged for new hardware.

    2) So please consider the users who want to build a new pc from scratch. They still have to get all those components. Motherboard, ram, cpu, etc. This is a more honest comparison really. And here are the facts:
    a) The Phenom II 940 is evenly priced with Core i7 920.
    b) The DDR3 1066RAM is evenly priced with DDR2 1066RAM
    c) The only more expensive component is the motherboard.

    And in return for spending an extra $100 for a motherboard, you get way better performance. Again, this holds true for a full new system.
     
  20. Iarwain

    Iarwain Banned

    Messages:
    3,047
    Likes Received:
    0
    GPU:
    4890 990//1120
    At the same time, why aren't you considering people who just want to upgrade? Your criticism falls against you as well. It sounds to me like you could've shortened the post if you had claimed "i7s are faster for the same price"

    I don't think anyone is disputing the i7's king of the hill here. Personally, I found it interesting that the AMD chipset showed huge advantages at high resolutions. I think you people are focusing too much on the processors themselves and you're ignoring that point. The fact of the matter is, that while the i7 may be the faster CPU, it is clearly paired with an inferior chipset. Judging from other benches I have seen, whatever motherboard the i7 is paired with doesn't matter, it's performance falls off drastically at high resolutions.

    This brings up the point I made earlier. There is something about the chipset, or even possibly the CPU, that is causing this falloff everyone is ignoring. In Hell's Highway at 1600x1200n the P2 drops 1 FPS. At the same resolution, in the same game, the i7 drops 9 FPS. What would happen if you stepped up again? What happens if you have an SLI/Crossfire configuration on an i7? Will you see worse scaling because of some oversaturation somewhere?

    I'm curious to know why no one is picking up on this. Everyone is anxious to say "AMD has a good processor, return to profit" or "i7 is the clear winner here, AMD has a loser processor" And yet there is a clear trend of huge performance drops when the GPU is really being stressed. This is worrisome to me. Ideas?

    It's possible this is just a flop, but the trend does seem to play out, to some extent, in all the benchmarks. It just strikes me that it actually falls below the P2, which indicates it's not specifically a GPU limitation.
     
    Last edited: Jan 17, 2009

Share This Page