Discussion AMD Ryzen MegaThread! FAQ and Resources

Page 30 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
I don't know what to believe anymore at the moment... What do you think about this review?

A 1700 neck to neck with 7700K@5ghz in gaming?? ?

https://www.youtube.com/watch?v=V5RP1CPpFVE

Could it be that after motherboards get optimised, Ryzen might become best CPU for content creation and gaming, bringing also the best bang for the buck? We can hope so I guess...
 


There are always reviews out there that don't jive with the rest of reviews, and then there are the photo chopped fake reviews that float around from time to time. To answer your question though, Ryzen just released today so nothing is optimized yet. AMD will be making optimizations and the motherboard manufactures will be making optimizations. There will be bios updates, there will be software updates, tweaks, in short everything that comes along with ironing out release day issues with a new die, silicon, and processor architecture. This is only the beginning for Ryzen, its not in its finalized polished stage. Intel's Kaby Lake is in a very refined polished stage as its in its 8th generation, but even Broadwell-E in 7th generation is shaken up by Ryzen in first generation outperforming it in almost every aspect, save 1080p gaming.
 


Just had a chance to review the video you posted, it is a legit review, however the reviewer tested everything in DX11. Where Ryzen has been coming up short is in 1080p DX12 testing. Most of the DX 11 tests show Ryzen neck and neck with Intel, and that is something that is also curious as DX11 in theory should have Intel way out in front and DX12 should have Ryzen closing the gap as DX12 allows for more cores to be utilized than DX 11 does. Further proof that there is some bug causing the sharp decline in performance in DX12 1080p gaming. Hopefully it is nothing more than a software related bug that can be fixed with an update or patch.
 
https://arstechnica.com/gadgets/2017/03/amd-ryzen-review/

Here is another review that shows Ryzen doesn't suffer much of a performance hit at all in 1080p gaming using DX11, in most titles. For some reason DX12, where Ryzen should in theory really shine, is where Ryzen takes the 1080p performance hit in gaming benchmarks.
 
Funny thing: Zen gets somewhere between 69.4% and 79.2% better single-core performance over Excavator in Cinebench R15, and somewhere between 10.7% and 23.4% better single-core performance over Excavator in Passmark. Guess what that averages out to?

Between 40.05% and 51.3% I think we may have found where AMD marketing was deriving their IPC gains.
 


Just to clarify Ryzen is NOT getting haswell level performance in gaming. Also i already gave a dolphin benchmark where Ryzen is 30% slower in IPC.

Something is odd with these results a lot of synthetic benchmarks show broadwell level IPC then games show between ivy and haswell, lets keep in mind the 6900K beats a 5Ghz 7700K in watch dogs 2 with Ryzen the 1800X is below the 7700K(at stock) by a decent margin. That means FAR below Haswell.

Something just seems off to me if it was truely at broadwell level then this processor should be keeping up with the 7700K quite easily in a lot of modern games as the 6900K keeps up with a 7700K in most modern titles despite having a weaker clock speed(also it has less IPC then the 7700K).

So no we need to keep this in mind a lot of gaming benchmarks is showing below haswell in IPC

https://youtu.be/TBf0lwikXyU?t=378

Gamer nexus talked to Amd 6:15 in the video live conversation EDIT AMD person claims in gaming in IPC broardwell and ryzen should be close???

Also in the beginning of the video Steve admits he was using the latest bios revision

It really does "grind my gears" how many times Amd told reviewers to test these chips to the point where the GPU is the bottleneck it really does.
 


Current version of AIDA64 doesn't read correctly RyZen parameters. Developers have even admitted that AMD didn't send them a sample.
 


Do someone really believes that AMD and game developers would do public statements about how will work together closely to improve the game performance of RyZen in future, if the gaming problem was solved with a simple BIOS update yesterday?

Joker? I mean really? The guy that cannot reproduce his own results due to large margin errors in his methodology? The guy that got the 1700 beating the 1800X? The guy that tested RyZen at 1080p but with odd settings that generated a GPU bottleneck with the GPU at 99% of load? The guy that has replied in tweeter to the criticism pretending that 99% of GPU load is not a bottleneck?
 


I know and i find him to be Amd bias(I mean he has a "Make Amd great again shirt") i could tell Steve from gamernexus was trying hard in their joint video to not correct him.

Joker was also corrected like several times by Jayztwocents, one time it was about something as simple as radiator mounting.

I'm subscribed to Joker but i think his reviews go with his name. Not to be mean or anything but man.

Juan just so you know its almost like Bulldozer all over again in the community(NOT THE PRODUCT)

People are saying "wait for moar cores in gaming", " "what about this magic patch/bios"
Something odd in synthetic benchmarks single threaded performance in Ryzen is good but in gaming its so so
 
One thing is that he was biased towards a given brand. Another very different is that he pretends he can fool us by saying that 99% load is not a GPU-bottleneck but a problem with v-sync

https://mobile.twitter.com/JokerReview/status/837356192716447751/actions

It doesn't matter if the problem is v-sync or if Mars was in conjunction with Venus. He tested in a GPU-bottleneck situation and thus the performance increase he reports was artificially created.
 


Lets hope they don't benchmark Ryzen with 3 4K monitors lol

I personally cannot understand why the tech community doesn't care that Amd is pulling these disgusting tactics, I know Intel wouldn't get away with this!
 


AMD pulling these tactics?

Care to elaborate? I already put a quote from the AMD AMA in Reddit where they stated they wanted a full mix, not one or another.

I haven't seen any hard evidence from anyone stating AMD was making reviewers create a bottleneck. Something like an email quote or picture of it. This is a "he said - she said" type of thing for me.

I do understand the need to test the CPU, but like I always say, more information never hurts. And to be totally fair, AMD gave reviewers too little time, so that can be the only thing I can say they did wrong in this day-1 review stuff.

Cheers!
 


If Intel told people to review their new kabylake processors in 4K or 1440P(GPU bound meaning the CPU is almost irrelevant) I can bet everything I own Amd fans would call them out and I will do the same with this. I find this to be pathetic.
I’d much rather have them be honest then I’d respect them but now they are making me respect them less
 
Yuka you listen to that video i showed? At this point Amd is pulling crap that i don't care for so happy Steve from gamer nexus called them out. If Intel pulled this the community would freak out.

Also disgusting tactics is me being nice when i'm talking about this!

EDIT

I hope mods don't think i'm trying to start something but i really do find this to be troubling to tell reviewers to test Ryzen in GPU bound situations seems to be anti consumer as once GPUs get more powerful 1440P results will look like 1080P results . Amd even went as far as lying or being or hiring people who are incompetent when talking about Ryzen's single core performance
 
My impression with this, my dear jdwii, is you're extrapolating a bit too much from a biased (to AMD) reviewer and mixing it with some reviewers interpretation of things without hard facts (like emails or proper quotes) backing their claims.

Don't let it get you too hard. Not worth it. I do agree it is troubling if you take it at face value, but I do wonder how they are interpreting AMDs words. I'm pretty sure Toms must have received some of those, but they haven't made a big deal out it.

If you ask me, the truth is always somewhere in-between, so in this case, AMD might have asked reviewers to *include* 4K or make it the *primary* source of conclusions and it got out of hand. You know how these things can spiral out of control, so try to take all options into account. Specially when it's a "he said - she said" type of thing.

Cheers!
 


By much less than that, the hype sites were filled one day before launch with headlines of accusations that Intel was trying to manipulate RyZen reviews. On the other hand, we have evidence and testimony from reviewers of AMD cheating demos and trying to manipulate reviews. A quote from french toast:

Ok i have to say this, we call intel out when do things like this , so fairs fair ill call amd just the same.

Not only did they mess up the reviews by not delaying, they did somthing much worse, they went all intel and tried to influence the reviews.

No i dont mean tips on their own products, i mean they asked several reviewers to change/disable settings for intel.
Such as, go into the command shell and turn off all core turbo for intel, run dual channel on intels, try to force 4k gaming comparisons to make ryzen look equal to intel, when they were behind (ironic because they didn't need to do that if they just waited for bios update).

In the space of 24 -36 hours they have effectively punched themselves in the face and lost alot of goodwill, stupid when they finally had a good product after 5 years.
Its this underhanded crap is the reason why i dont follow certain companies, amd better not turn into one of them.

I still have sypathy for their cause, just as long as they dont start acting like intel.

 


I tend to agree with Yuka here. Why wouldn't AMD want its new high end processor tested at 1440p and 4K resolution? Isn't that the apex of gaming right now? It's something its previous flagship FX Piledriver processors couldn't do, AMD wanted to illustrate its new Ryzen processor could. I mean seriously we debate things like the GTX 1080, and 1080Ti and AMD Vega which is the best enthusiast GPU, why would it even matter if people weren't buying them? Does anyone really believe that people are going to buy them and not play games at 1440p and 4K resolutions? Why would you pay that much for a GPU if your going to play at 1080p when there are a lot of much cheaper GPUs that can run 1080p? Obviously people buying the 1080, 1080p, AMD Fury X, AMD Vega are interested in enthusiast 1440p and 4K gaming resolutions- so why wouldn't AMD want to show how its new enthusiast market processor games at 1440p and 4K resolutions?

I think AMD was so concerned with Vega vs 1080Ti and Kaby Lake vs Ryzen at 4K that they missed something at 1080p resolution using the newer DX12 API. If you look at a lot of benchmarks out there Ryzen trades blows with similar Intel processors (I'm talking the i7 6800K, 6900K) in a lot of DX11 API games. Yet in those same games that also allow for DX12 API all of a sudden Ryzen takes a significant hit in FPS. Ryzen in DX12 should perform much better than in DX11 and in a lot of titles its either a minor bump or its the exact opposite. Rather than a problem with the architecture I think its a software related issue which is why in benchmarks (single core and multi-core) Ryzen is toe to toe or better than i7 6900K but in actual DX12 titles Ryzen is struggling where is should be dominating. Its also worth mentioning that in games that utilize Vulkan Ryzen does much better. All these facts point not to a major architectural problem but a software problem where Ryzen isn't utilizing DX12 to its fullest potential. In DX 11 we should see Intel leading across the board with slightly better IPC and higher clock speeds but in most titles Ryzen keeps its close. Where Ryzen should shine is DX12 where more cores equals more gaming capability but we are in fact seeing the exact opposite the majority of the time. Its almost like something with DX12 is making the processor fight itself and take a significant performance hit.
 
As far as AMD asking Intel to disable certain aspects of Intel processors for testing basically hobbling them, I've got nothing. I hate when Intel does that so I'm will not defend AMD for trying the same thing. Yea, turn about is fair play, but not when its so easily discovered and not when your trying to make a case as the "clean" company up against what we all know Intel to be. I can understand AMD wanting its processors compared to Intel in 1440p and 4K gaming with both processors running as optimal as they can, nothing more and nothing less. I understand that because 1440p and 4K is the new enthusiast gaming levels and a lot of computer gamers are buying the high end GPUs to run at those levels. AMD was caught with their pants down in the 1080p DX12 benchmarks which needs to be addressed right away before the "Ryzen is a a failure hype train" gains any more momentum than it already has garnered.
 


I think we have to be very careful as a community to make sure that a "Ryzen is a failure hype train" doesn't get perpetuated by respected experts here at Tom's Hardware. Tom's is one of the most respected tech sites there is, which is why I like to come here and contribute when and where I can. By dwelling on noting but Ryzen's one shortcoming (which to be honest isn't as huge of a problem as it looks as even with its current problem its still a very capable gamer (unlike Bulldozer was)) and not even discussing Ryzen's strong points we are adding fuel to that hype train.

Yes, Ryzen has a problem with 1080p gaming as of right now, especially utilizing the new DX12 API. Its not out of the realm of possibility that a bios update and software optimizations won't help to fix the issue or that DX12 isn't reading the new AMD chip properly which could also be fixed via patch. Long story short ,the problem is there, AMD is aware of the issue and I'm sure they are burning the midnight oil to fix it. I don't believe its all as simple as software (ie games) need to be optimized for the new architecture, but I also don't believe its the architecture itself so updates and patches that I'm sure will be incoming should help.

In the meantime there is a huge growing "failure hype train" that is there and growing, one that is having a lot of fuel added to the fire because no one is talking about Ryzen's many strengths only its one glaring weakness. Correct me if I'm wrong, but I can remember you Juan stating way back before DX12 released that it wouldn't be a "magic bullet" for FX Piledriver and even if it somehow was wouldn't help AMD regain footing in the server market where all the real money is and where AMD really needs to get back into. Now that AMD has a processor capable of doing exactly that I haven't really seen you talking about the server market or AMDs upcoming Zen server processors. The R7 1800X is thrashing the much more expensive i7 6900K in workstation based loads. AMD has the processor to re-enter the most lucrative market there is, yet all we talk about is 1080p gaming, a market you pointed out isn't anything when compared to the server market. All the tech sites are focusing on the same thing, 1080p DX12 gaming and vastly ignoring the fact that Zen is going to regain huge portions of the server market with its raw workstation performance. We are scarring investors so bad that stock has fallen for the last 2 days and will continue to fall all for an issue that isn't as great as its being blown out of proportion to be. This isn't Bulldozer all over again, Bulldozer couldn't effectively game without bottlenecking games with more powerful GPUs, something that Ryzen doesn't have an issue with. Bulldozer never had the possibility of regaining significant market share in the server market, Ryzen has a legitimate chance to do so based on its workstation prowess. Yes we should discuss the 1080p DX12 gaming problem, but we have to keep it in perspective so as not to scare the living heck out of investors, which AMD desperately needs to be able to iron out all of Ryzen's performance potential.
 
http://www.game-debate.com/news/22425/over-300-plus-developers-are-working-to-optimize-amd-ryzens-performance-on-their-games

AMD is taking the problem seriously, hopefully by the time the R5 releases 1080p gaming issues will be addressed. I find the statement "In regards to poor competitive performance using SMT, Dr. Lisa said that some games are using code which is optimized for their rivals and hence Ryzen is suffering from performance dips" to be interesting as I can remember a time when Intel built in a "cripple AMD code" to their compiler that was vastly used in lots of applications including benchmark software that landed Intel in court. This is a legitimate concern as its not a "conspiracy theory" if someone has done something exactly like that before and we are seeing very similar again. Just something to keep an eye on, although in no way a smoking gun.
 


Don't worry, there are so many mud thrower or other that say like or dislike technology but not companies, any reference is totally random, that will keep not reading or pretending that this is not clear to them. I tried to explain them from the bottom of my noobness but my other things deeper comprehension then others, marketing, strategy, finance, business, sales (and learning from what others have wisely, like you, pointed out) that this is a process, it's not a product. There are even people that believe that security is a product.
Yes ... it is a product of a process!

So will be everything good you wanna see. Ryzen has ryzed... gamers I repeat, make a lot of noise from the bottom of their place (I am sorry but average player is there - it's statitical, it's a fact, because ppl r often "too young") and so on internet you will see these "lamers" taking over Ryzen as a failure and all the other people that have no critical sense (most of us) or indipendence in creating a structured opinion to believe that Ryzen is not worth it.
Hopefully some good journalism (almost absent today) and good reviers will keep opening the eyes of people despite everything.