AMD responds to 1080p gaming tests on Ryzen. Supports ECC RAM. Win 10 SMT bug

[zMeul]

1 minute ago, leadeater said:

This is it though https://www.extremetech.com/computing/220953-skylake-bug-causes-intel-chips-to-freeze-in-complex-workloads, timing of that update make me think that was the fix.

that's the one

some mobo manufacturer's issued micro-code updates since January, others did in March

[MoraisGT]

People calling Ryzen a fail because it does not perform as good as a 7700K in games should outright refrain themselves from commenting on this forum.

 

The Ryzen 7 CPU lineup was never designed to good for gaming, it's all about hevy multithreaded applications. And it excels at that being much cheaper than Intel's alternatives.

 

But even for gaming, they are very capable CPU's.

Chech out this comparisson between a 1700 @ 3.9GHz and a 7700K @ 5GHz at 1080p:

 

 

 

Let's face the facts, the performance is incredible, even more so considering their last lineup of CPU's.

[Morgan Everett]

" Ryzen is doing really well in 1440p and 4K gaming when the applications are more graphics bound".

 

Oh, dear. 

[VagabondWraith]

15 minutes ago, Morgan Everett said:

" Ryzen is doing really well in 1440p and 4K gaming when the applications are more graphics bound".

 

Oh, dear. 

"My Titan X gets 10,000 FPS at 720p everything maxed out."

[Memories4K]

15 minutes ago, MoraisGT said:

People calling Ryzen a fail because it does not perform as good as a 7700K in games should outright refrain themselves from commenting on this forum.

Let me just help you here by saying maybe 10% of people(arbitrary number based on nothing, i'm saying an extremely small minority), TOPS, actually are disappointed it does worse than a 7700k.

Most people are disappointed because the 1800X seems to be so close to the 6900k in synthetic benchmarks and production work (which is great for it's value, being half the price) but has a MUCH bigger difference in games.




The question people are asking is why is the gaming performance so dramatically different, when in other areas the 1800X can keep up and at times even beat the 6900k? (Keep in mind that the 6900k is Broadwell-E)

The 7700k comes up in the discussion only when somebody asks "Why would i pay more money for a $330/$380/$500 CPU over a $350 7700k or $240 7600k that does gaming better for cheaper?" which i haven't seen too often as most people i think already knew the 7700k would outperform it but in cases where the 7700k does outperform the 1800X it's by a MUCH wider margin than the difference between the 7700k and the 6900k. Something else is that when the 6900k actually matches or overtakes the 7700k, the difference is STILL there and the 1800X STILL lags far behind, which is interesting considering the 7700k does have faster single-threaded performance and higher clocks than the 6900k which must mean that the extra threads in the 6900k must be being utilized to the point it CAN match/jump over the 7700k, where the 1800X isn't seeing similar benefit.

Now looking at the benchmarks i think Watch Dogs 2 is a good example, even with SMT off.
BF1 is actually a lot closer but notice the difference between SMT on (stock) vs SMT off (stock).
You keep looking at these benchmarks and something is very wrong, i don't believe the 1800X is being utilized properly at all.
You're going to tell me that the same CPU @ stock that can keep up with the 6900k in other benchmarks/productivity work is now suddenly being beat by a 6900k @ stock when that same CPU is overclocked?

[MoraisGT]

1 hour ago, Memories4K said:

Now looking at the benchmarks i think Watch Dogs 2 is a good example, even with SMT off.
BF1 is actually a lot closer but notice the difference between SMT on (stock) vs SMT off (stock).
You keep looking at these benchmarks and something is very wrong, i don't believe the 1800X is being utilized properly at all.
You're going to tell me that the same CPU @ stock that can keep up with the 6900k in other benchmarks/productivity work is now suddenly being beat by a 6900k @ stock when that same CPU is overclocked?

 

It is very tricky to explain indeed.

 

We have also seen very disparing results in the same benchmarks from different reviewers. 

This tells me that the BIOS's are probably not yet up to the task. An example of this is TTL from OC3D, where during his testing he could not get past 2666MHz in RAM, but after he did his review, a new BIOS came along and...easy 2933MHz RAM all of the sudden.

 

But I also believe that what AMD says has some truth to it. Game developers only had to worry about Intel chips for the past half-decade, so maybe you could see improved performance in a few months with patches along with new BIOS revisions?

 

We'll have to wait and see. I for one am very happy that AMD after such a long time was able to pull this off, I mean a 329$ chip (with a minor OC) that can rival a 1000$ chip from Intel in multithreaded workloads? 

 

*Clap* *Clap* AMD

[pas008]

2 hours ago, MoraisGT said:

People calling Ryzen a fail because it does not perform as good as a 7700K in games should outright refrain themselves from commenting on this forum.

 

The Ryzen 7 CPU lineup was never designed to good for gaming, it's all about hevy multithreaded applications. And it excels at that being much cheaper than Intel's alternatives.

 

But even for gaming, they are very capable CPU's.

Chech out this comparisson between a 1700 @ 3.9GHz and a 7700K @ 5GHz at 1080p:

 

 

 

Let's face the facts, the performance is incredible, even more so considering their last lineup of CPU's.

say what never designed for gaming but all the hype, news, etc for last 6months were just after production and not gamers?

 

but on side note I do see alot of inconsistencies too even in some synthetic benches

leaves to to wonder if bios, drivers, os fix, will make them surpass what all we have seen

this is amd though they fix after release lol

[Trixanity]

I have wondered whether February's Windows patches were delayed because a Ryzen patch broke something. We know it was delayed because there were unexpected issues. Perhaps there is a connection. I know it's a long shot. We'll see if there is any merit to that in about two weeks.

And I also realize that there are many more problems to solve than that, so it's not like I'm saying "this patch is all that is needed and everything will be fine".

 

AMD is obviously working on microcode updates, BIOS updates with motherboard vendors, OS developers and application/game developers to get everything up to speed. This will take some months. By then things will become more clear about where Ryzen stands when the dust has settled. I'm sure many of these optimizations will also play a part in Naples which need to be rock solid including the alleged memory issues. So you could argue that current Ryzen owners are kinda beta testing. By the time Naples and R5 launches (and hits a much wider audience) the worst of it should have been solved (if it can be solved).

 

Should AMD have waited? No: AMD couldn't delay further. They needed the money and they had to get the ball rolling. A delay would have been worse both financially (imagine their stock if they delayed) and competitively (no CPUs on the market worth talking about). The chip works but it needs more polishing. They can do that even though it looks bad.

[DragonTamer1]

48 minutes ago, pas008 said:

say what never designed for gaming but all the hype, news, etc for last 6months were just after production and not gamers?

 

but on side note I do see alot of inconsistencies too even in some synthetic benches

leaves to to wonder if bios, drivers, os fix, will make them surpass what all we have seen

this is amd though they fix after release lol

Pretty much all of those patches need to happen. Was watching tech talk last night and Jay reaffirmed all of it. He also talked about what was going on behind the scenes of the reviewers.

 

Basically the BIOSs' that are being shipped are trash because the board partners didn't have a lot of time to refine what they had.

 

Windows 10 needs optimization s before it can fully leverage all of the Ryzen CPU. And the game developers need to tweak their code to work with the new architecture.

 

Basicallt it was a rushed launch and everyone was caught with their pants down.

[MoraisGT]

Good video on this topic.

 

 

[Ryujin2003]

So... When do we start boarding for the R5 1600x Hype Train?

[done12many2]

Just now, MoraisGT said:

Good video on this topic.

 

 

 

Very good video.  I'm just at a loss for the lack of professionalism with this release.  Everything from lack of preparation to intentional shaping of outcomes.  Way too much influence or oversight on reviews.

[Memories4K]

Just now, done12many2 said:

 

Very good video.  I'm just at a loss for the lack of professionalism with this release.  Everything from lack of preparation to intentional shaping of outcomes.  Way too much influence or oversight on reviews.

Steve holds to his own integrity and i find that very admirable.

[done12many2]

Just now, Memories4K said:

Steve holds to his own integrity and i find that very admirable.

 

Yeah, I held Jayz video about AMD's influence with a grain of salt as I'm out of sync with his views at times, but when I saw GN (Steve's) I now realize how bad this oversight/influence from AMD actually was.

[Guest]

3 minutes ago, done12many2 said:

 

Very good video.  I'm just at a loss for the lack of professionalism with this release.  Everything from lack of preparation to intentional shaping of outcomes.  Way too much influence or oversight on reviews.

For real. I didn't expect AMD to outdo Intel.

 

But I certainly didn't expect them to pull so much marketing bullshit. 

 

I mean, some of these practices are straight up unethical. 

[Tomsen]

5 minutes ago, Alexokan said:

For real. I didn't expect AMD to outdo Intel.

 

But I certainly didn't expect them to pull so much marketing bullshit. 

 

I mean, some of these practices are straight up unethical. 

Can you name a few of those practices?

[Valentyn]

12 minutes ago, MoraisGT said:

Good video on this topic.

 

 

Same issue Golem.de had. Their MSI board on BIOS 113 was terrible, where as on 117 they saw very good performance increases, and stability.

 

Motherboard manufactures just really need to sort out the issues hampering performance, but then again it really is AMD's fault for rushing out the launch

 

https://translate.google.co.uk/translate?hl=en&sl=de&u=https://www.golem.de/news/ryzen-7-1800x-im-test-amd-ist-endlich-zurueck-1703-125996-4.html&prev=search

 

Quote

The MSI board was delivered with BIOS version 113, until last Friday a new one appeared.

Version 117, which is still up-to-date, improved speed and stability. If we were still able to count on sporadic Bluescreens with the older UEFI, the board is currently stable. Much more important, however, is the drastically higher performance in games and the real pack with 7-Zip. The release notes include, among other things, a fixed problem with the memory act and its timing as well as the voltage.
Compared to the original bios, the new UEFI increases the image rate in our game course between plus 4 and plus 26 percent, on the average even plus 17 percent!

 

 

Could explain why reviewers with the Gigabyte motherboards, like Joker, and Crit got much better gaming numbers.

[Tomsen]

1 minute ago, Valentyn said:

Motherboard manufactures just really need to sort out the issues hampering performance, but then again it really is AMD's fault for rushing out the launch

What makes you think AMD rushed it? IIRC motherboard manufacturers have had ryzen CPU for more than a few months prior to launch. Who is to blame?

[Valentyn]

6 minutes ago, Tomsen said:

What makes you think AMD rushed it? IIRC motherboard manufacturers have had ryzen CPU for more than a few months prior to launch. Who is to blame?

 

I get what you mean, but it's two way street really. When it came to AIDA64, AMD never even provided them with any hardware, and they still don't have any.
So their benchmarking and software doesn't know how to handle the architecture either.

 

What this all does mean to me, is that there's more performance to be gained. Especially if a last minute MSI BIOS update can net an average increase of 17% in performance.

[Memories4K]

9 minutes ago, Tomsen said:

Can you name a few of those practices?

Steve was told by an AMD representative to create a GPU bottleneck in order to close the performance gap and Jay was told to disable certain Intel features in his review to reduce Intel CPU performance.

[Tomsen]

9 minutes ago, Valentyn said:

 

I get what you mean, but it's two way street really. When it came to AIDA64, AMD never even provided them with any hardware, and they still don't have any.
So their benchmarking and software doesn't know how to handle the architecture either.

 

What this all does mean to me, is that there's more performance to be gained. Especially if a last minute MSI BIOS update can net an average increase of 17% in performance.

Well, AIDA64 is a benchmark, not a motherboard manufacturer. You don't necessarily have to provide benchmark manufacturers with hardware, since changes to the compiler in use should be able to handle that. 

 

7 minutes ago, Memories4K said:

Steve was told by an AMD representative to create a GPU bottleneck in order to close the performance gap and Jay was told to disable certain Intel features in his review to reduce Intel CPU performance.

Steve who? How did he get told? What exactly did he get told? Source?

Jay who? How did he get told? What features were he suppose to disable? Source?

[done12many2]

1 minute ago, Tomsen said:

Well, AIDA64 is a benchmark, not a motherboard manufacturer. You don't necessarily have to provide benchmark manufacturers with hardware, since changes to the compiler in use should be able to handle that. 

 

Steve who? How did he get told? What exactly did he get told? Source?

Jay who? How did he get told? What features were he suppose to disable? Source?

 

The videos for both are already in this thread.

[Dietrichw]

16 minutes ago, Memories4K said:

Steve was told by an AMD representative to create a GPU bottleneck in order to close the performance gap and Jay was told to disable certain Intel features in his review to reduce Intel CPU performance.

That is taking things a bit out of context (or can be taken out of context, that is how wild rumors start), they did not ask for a GPU bottleneck to be created for all tests. They wanted higher res benchmarks included along with 1080p, not GPU bottlenecked tests only. Jay was asked to disable a feature that showed up on later X99 boards which helps the chips reach single core  turbo across all the cores which is a sort of auto overclock which isn't really the stock performance of X99 chips but is on by default on every new X99 board nowadays. Since it such a default feature Jay has a good reason to not disable it as no real user will disable it or know it even exists, but it is a feature that would not be found on an original launch X99 board or an Intel made motherboard.

[Tomsen]

Just now, done12many2 said:

 

The videos for both are already in this thread.

We are 7 pages deep, and I see a bunch of videos posted. Could you do me the favor or linking to them (or the comments), so I don't have to go through all the pages and videos.

[Memories4K]

2 minutes ago, Tomsen said:

Steve who? How did he get told? What exactly did he get told? Source?

Jay who? How did he get told? What features were he suppose to disable? Source?

Steve, from GamersNexus [Source] Over the phone/e-mails

Jay, from JayzTwoCents [Source] Over e-mails