Der8auer: "Good number of Ryzen 3000 chips does not reach boost clock"

I had a feeling it was wrong, digging up their latest beta shows proper results now.

 

7nm itself is still quite new process node, maybe matured one produce better chips time will tell

back on release i read this : https://www.techpowerup.com/forums/...lf-the-performance-on-single-ccd-cpus.257182/
when i read that, as far as performance didnt think it effect much whatsoever, so never dig more about the issue

but then we read the news mobo maker limiting boost few weeks ago, asus said for longevity ? (https://www.guru3d.com/news-story/r...ior-in-favor-of-longevitysays-asus-staff.html)
so i wonder if all of this related to mobo, chipset, bios, rather than the cpu itself? or its really the cpu limitation ?

well there is no (human-made) product perfect anyway,
for ryzen, imo overall the value is still great with its pricing

 

TBH I would have never even noticed that,I only ever looked at the sensors.

I am a AMD fanboy and just a couple post up I wrote

not sure where why you would need to state everyone wants your money of course they do otherwise the world might be a better place kumbaya MF lol .

It seems and I would like to ask so when did AMD Ryzen get so slow that it is trash now,from what I recall I seen most all reviews the Ryzen 3 CPU lineup for cheaper than Intel still beating Intel.Intel lowered prices and I would still pick AMD over Intel.
The shame for AMD your slow 4300Mhz CPU beating Intel 5100Mhz Intel ,shame on you AMD,Shame on you.

 

Correct - Intel was caught sleeping.

 

Yes you are!

Correct - Intel was caught sleeping on this one.

P.S. Sorry for the double post - an admin please delete the post before this...

 

Tbh the results shouldn't be taken seriously when you have reddit posters boasting that submitted multiple times low speed values to get to AMD. There is no control on the data sample.

My 3900X does 4525 out of the box without tweaking. When turn on game mode on Ryzen master (effectively make it 6c/12t) the boost goes to 4.75 on CCX1. And that with a NH D15

 

Sadly on this test we have no idea if the cpu cooler was properly installed and what motherboard was used. Does the series x300 boost as good? ( i remember they do not have xrf2 ) also the point on whatever else is running on windows background is valid.

 

You get what you pay for.

Pay less, getting less

Still happy owner of 3900x

 

Why would you say that???
Yes, the frequency is variable across platforms and users but still the performance is great and the new series just oozes value compared to competition.

 

AMD ain't saint, everyboddy should know this, my buying decisions are not influenced by that. That said, I belieave it's a minnor problem, but shouldn't happen anyway. If all processors can only boost to 4.2 so advetise 4.2, it's simple, there is no escuse. Ryzen 3k still a great buy anyway (despite AMD not being saint).

 

Pentium G5400, £57, happy Pentium owner

 

Another rant fanboy ism

Doesnt matter if you favour one over the other, some stupid remarks made - Amd has come a long way, and intel fannies can now respect Amd pussies

Amd has now set Intel on fire and with that ! Its good for everybody , we all come out winners, if you cannot see that, go to spec savers

 

Im skeptical of the software reporting, main reason being is that these cpus change clock frequency faster than any software polling could accurately detect, if i recall ryzen uses a 1 mhz rate for clock adjustments.

unless amd if doing some kind logging on chip, I don't see how it could be accurate(for measuring peak boost anyway)

 

My 3900X does not want to hit the advertised boost clock, but then again, I run my ram and IF at 3733mhz with 16-15-15-32 1T timings (AIDA reports 64.1 ns latency) and from what I understand, increasing the IF speed potentially decreases the boost. Does not really matter to me though, I am running all core 4.35Ghz @ 1.35v 100% stable. 7756 C20 multi score. 4.4Ghz all core is a no go stable though, at least not without WAY higher voltages than I'd want to do 24/7 to get it totally stable.

 

Just use task manager
https://i.**********/jdvQgNtq/gg2.png

 

1. You get the performance with the lower clocks. So? 3600 is 4,2 boost, 3600X is 4,4 boost? And what's the difference in their performance? Near zero. So what random performance are you lying about?
2. Hyped for the nex X570 platform which gives zero performance gain, but gives you a motherboard fan for extra noise? Lol

Ahahaha, you are a funny guy. Intel officially did that by saying their CPUs are still better in real life applications by showing only 1 example of those (in 3 cases) and making averages out of those 1-1 results on their graph.

 

Don't care about the whole "how manys GHZ" thing, bottom line is, do your research, check benches, reviews etc. My new Ryzen actually satisfied my expectations performance wise, despite it not reaching so called claimed max ghz boosts. I feel like i got a good value in it.

 

It seems to me that this is waaaaaay less of a big deal than some are making it out to be. A couple hundred mhz, a few hundred points one way or another on a benchmark........this isnt all that uncommon. When has it not taken a few steppings and a few updates to get a new chip architecture up to snuff? Of course they want to spin their numbers as positive as possible, that is how marketing works. Good, bad, ugly, or indifferent, that is how the game is played, in every business. Most of all in global scale businesses. 100% of all companies want you to spend your money with them, not their competitors. None of this is any reason to be insulting to one another......it just makes us IT folk/computer nerds look like dumb jerks. By and large, we are not.

 

So if a box of cookies said it came with 12 and you only got 9, you'd be okay with that?
Or if you are out to dinner and the menu said you get fries with your meal and they don't give you the fries, you'd be okay with that too?

You are clearly the type of customer these types of businesses want, because you don't mind getting ripped off, apparently.