Discussion AMD Ryzen MegaThread! FAQ and Resources

[gamerk316]

What about the 250ms hang in frame times from the 7700K that his graph showed? The 5960X also experienced a hang that was 180ms.

Considering that Ryzen never broke 30ms gap, and at no point was considered stuttering, I would think the hard numbers from the frame time graph would corroborate the user experience. Especially true if we are considering it was duplicated by multiple individuals playing the same games across the same systems....

There is your definitive proof that Ryzen was smoother than Intel. Frame times do not lie, and a 180-250ms hang in frame times would be a horrendously bad stutter with a noticeable effect on performance.

I'm going to point out that normal tech review sites have found absolutely no evidence of those types of hangs, especially when you consider that a 250ms hang time represents 16 frames being consecutively dropped. Which would be quite noticeable, even without latency benchmarking.

So I'll come out and say it: the test is being rigged is some way to make Intel appear worse. Because if the 7700k (or other, slower, Intel CPUs such as my 2600k) had those types of issues in normal usage, someone would have noticed by now. Or are you going to accuse every single tech site in existence of being Intel shills for not reporting this issue?

Gamer, he has the fraps capture from the hang, and it was run during the video.

Seriously...the test is not being rigged. He even comments that he would have thought Nvidia's drivers should have just dropped the single frame and moved on, it makes the most sense from his perspective. The drivers, however, did not do that. He checks signal interrupts and all sorts of things. If you develop on windows, he literally had EVERYTHING perfmon checks enabled to see what was going on the whole time.

I note FRAPS is very CPU intensive. It's a useful tool for getting FPS out of the GPU, but you are going to have latency hits, especially if Windows stuffs it on a Hyperthreaded core. i mean seriously, how many posts do we have daily about people complaining about shoddy gameplay while FRAPS is up and running? If you have a heavy CPU hog running in the background, CPU performance suffers. And by virtue of having more cores, Ryzen handles this case better.

Nothing to see here.

 

[8350rocks]

What about the 250ms hang in frame times from the 7700K that his graph showed? The 5960X also experienced a hang that was 180ms.

Considering that Ryzen never broke 30ms gap, and at no point was considered stuttering, I would think the hard numbers from the frame time graph would corroborate the user experience. Especially true if we are considering it was duplicated by multiple individuals playing the same games across the same systems....

There is your definitive proof that Ryzen was smoother than Intel. Frame times do not lie, and a 180-250ms hang in frame times would be a horrendously bad stutter with a noticeable effect on performance.

I'm going to point out that normal tech review sites have found absolutely no evidence of those types of hangs, especially when you consider that a 250ms hang time represents 16 frames being consecutively dropped. Which would be quite noticeable, even without latency benchmarking.

So I'll come out and say it: the test is being rigged is some way to make Intel appear worse. Because if the 7700k (or other, slower, Intel CPUs such as my 2600k) had those types of issues in normal usage, someone would have noticed by now. Or are you going to accuse every single tech site in existence of being Intel shills for not reporting this issue?

Gamer, he has the fraps capture from the hang, and it was run during the video.

Seriously...the test is not being rigged. He even comments that he would have thought Nvidia's drivers should have just dropped the single frame and moved on, it makes the most sense from his perspective. The drivers, however, did not do that. He checks signal interrupts and all sorts of things. If you develop on windows, he literally had EVERYTHING perfmon checks enabled to see what was going on the whole time.

I note FRAPS is very CPU intensive. It's a useful tool for getting FPS out of the GPU, but you are going to have latency hits, especially if Windows stuffs it on a Hyperthreaded core. i mean seriously, how many posts do we have daily about people complaining about shoddy gameplay while FRAPS is up and running? If you have a heavy CPU hog running in the background, CPU performance suffers. And by virtue of having more cores, Ryzen handles this case better.

Nothing to see here.

Why did the OC'ed 5960X also hang?

 

[8350rocks]

https://www.youtube.com/watch?v=kKUV74lL9DA

PS2 emulation on PC + streaming at the same time on Ryzen 1700

 

[James Mason]

https://www.youtube.com/watch?v=kKUV74lL9DA

PS2 emulation on PC + streaming at the same time on Ryzen 1700

That's not... news?

 

[jdwii]

https://forums.dolphin-emu.org/Thread-ryzen-1700x-stock-vs-dolphin-benchmark

Not to bad 10% lower then a 4770K meaning in this case Ryzen is below Haswell IPC but not by a massive margin

Better benchmark
http://www.legitreviews.com/wp-content/uploads/2017/03/dolphin-1700.jpg


From that dolphin thread i found this
https://www.techpowerup.com/img/17-03-06/0ad6052cf561.jpg

 

[-Fran-]

https://forums.dolphin-emu.org/Thread-ryzen-1700x-stock-vs-dolphin-benchmark

Not to bad 10% lower then a 4770K meaning in this case Ryzen is below Haswell IPC but not by a massive margin

Better benchmark
http://www.legitreviews.com/wp-content/uploads/2017/03/dolphin-1700.jpg


From that dolphin thread i found this
https://www.techpowerup.com/img/17-03-06/0ad6052cf561.jpg

For the second image, the first two columns are: "Read" and "Write" respectively.

Cheers!

 

[gamerk316]

What about the 250ms hang in frame times from the 7700K that his graph showed? The 5960X also experienced a hang that was 180ms.

Considering that Ryzen never broke 30ms gap, and at no point was considered stuttering, I would think the hard numbers from the frame time graph would corroborate the user experience. Especially true if we are considering it was duplicated by multiple individuals playing the same games across the same systems....

There is your definitive proof that Ryzen was smoother than Intel. Frame times do not lie, and a 180-250ms hang in frame times would be a horrendously bad stutter with a noticeable effect on performance.

I'm going to point out that normal tech review sites have found absolutely no evidence of those types of hangs, especially when you consider that a 250ms hang time represents 16 frames being consecutively dropped. Which would be quite noticeable, even without latency benchmarking.

So I'll come out and say it: the test is being rigged is some way to make Intel appear worse. Because if the 7700k (or other, slower, Intel CPUs such as my 2600k) had those types of issues in normal usage, someone would have noticed by now. Or are you going to accuse every single tech site in existence of being Intel shills for not reporting this issue?

Gamer, he has the fraps capture from the hang, and it was run during the video.

Seriously...the test is not being rigged. He even comments that he would have thought Nvidia's drivers should have just dropped the single frame and moved on, it makes the most sense from his perspective. The drivers, however, did not do that. He checks signal interrupts and all sorts of things. If you develop on windows, he literally had EVERYTHING perfmon checks enabled to see what was going on the whole time.

I note FRAPS is very CPU intensive. It's a useful tool for getting FPS out of the GPU, but you are going to have latency hits, especially if Windows stuffs it on a Hyperthreaded core. i mean seriously, how many posts do we have daily about people complaining about shoddy gameplay while FRAPS is up and running? If you have a heavy CPU hog running in the background, CPU performance suffers. And by virtue of having more cores, Ryzen handles this case better.

Nothing to see here.

Why did the OC'ed 5960X also hang?

My guess? A major thread is getting assigned to a HTT core, and stalling out. Hence the downside to adding more and more threads: A greater and greater chance of a major program thread getting assigned to a SMT core and stalling out. I could confirm this if I had the raw perfmon logs, but based on observation and known Intel CPU performance, this explanation makes the most sense.

My point here is the test itself is invalid: You can't try and gauge CPU performance within an application while at the same time having another CPU intensive program running in the background.

 

[juanrga]

https://forums.dolphin-emu.org/Thread-ryzen-1700x-stock-vs-dolphin-benchmark

Not to bad 10% lower then a 4770K meaning in this case Ryzen is below Haswell IPC but not by a massive margin

Better benchmark
http://www.legitreviews.com/wp-content/uploads/2017/03/dolphin-1700.jpg

I would ignore the first, because it is from one of those fake leaks released before launch. The performance reported in that leak wasn't confirmed by reviews.

And I would ignore the second because it is internally inconsistent. The 1700 and the 1800X are the same chip except for clocks, and the graph says that the 1700 OC to 4GHZ on all cores is slower that the 1800X at stock with 3.6GHz base and single core 4GHz turbo. A possible explanation of the discrepancy is that the 1800X is running with XFR activated in whose case the effective clocks aren't the clocks reported in the graph.

Here you have Dolphin benches from computerbase

Zen @3.6GHz performs like Haswell @3GHz in this bench.

 

[cdrkf]

https://forums.dolphin-emu.org/Thread-ryzen-1700x-stock-vs-dolphin-benchmark

Not to bad 10% lower then a 4770K meaning in this case Ryzen is below Haswell IPC but not by a massive margin

Better benchmark
http://www.legitreviews.com/wp-content/uploads/2017/03/dolphin-1700.jpg

I would ignore the first, because it is from one of those fake leaks released before launch. The performance reported in that leak wasn't confirmed by reviews.

And I would ignore the second because it is internally inconsistent. The 1700 and the 1800X are the same chip except for clocks, and the graph says that the 1700 OC to 4GHZ on all cores is slower that the 1800X at stock with 3.6GHz base and single core 4GHz turbo. A possible explanation of the discrepancy is that the 1800X is running with XFR activated in whose case the effective clocks aren't the clocks reported in the graph.

Here you have Dolphin benches from computerbase

Zen @3.6GHz performs like Haswell @3GHz in this bench.

Is that a single thread test per chance?

Edit: there is something strange about the scaling of that. If you look at the 1700- it's above a 3770k, which if we are talking base clocks has Ivy at a 500mhz advantage (3.5ghz base on the 3770 vs 3 on the 1700). If we are looking at turbo clocks then it's 3.9ghz vs 3.7ghz, however then your assertion that 'Zen @ 3.6ghz = Haswell @ 3ghz is wrong', because the 5960x turbos to 3.5ghz (so it would be 4 vs 3.5). The only thing that is strange about that though is the huge gap between Ivy and Haswell in this test, is this an AVX 2 enabled benchmark? Haswell didn't really have much ipc over Ivy in most test from what I remember...

 

[XBloodyR]

@cdrkf

Its a synthetic benchmark thats why you cant really take it seriously. Probably most of what he has postet are synthetic ones.

 

[James Mason]

Also that benchmark looks extremely single threaded, so I imagine if there were any i3s on there they'd be just under the i7s and still above Ryzen.

 

[juanrga]

https://forums.dolphin-emu.org/Thread-ryzen-1700x-stock-vs-dolphin-benchmark

Not to bad 10% lower then a 4770K meaning in this case Ryzen is below Haswell IPC but not by a massive margin

Better benchmark
http://www.legitreviews.com/wp-content/uploads/2017/03/dolphin-1700.jpg

I would ignore the first, because it is from one of those fake leaks released before launch. The performance reported in that leak wasn't confirmed by reviews.

And I would ignore the second because it is internally inconsistent. The 1700 and the 1800X are the same chip except for clocks, and the graph says that the 1700 OC to 4GHZ on all cores is slower that the 1800X at stock with 3.6GHz base and single core 4GHz turbo. A possible explanation of the discrepancy is that the 1800X is running with XFR activated in whose case the effective clocks aren't the clocks reported in the graph.

Here you have Dolphin benches from computerbase

Zen @3.6GHz performs like Haswell @3GHz in this bench.

Is that a single thread test per chance?

Edit: there is something strange about the scaling of that. If you look at the 1700- it's above a 3770k, which if we are talking base clocks has Ivy at a 500mhz advantage (3.5ghz base on the 3770 vs 3 on the 1700). If we are looking at turbo clocks then it's 3.9ghz vs 3.7ghz, however then your assertion that 'Zen @ 3.6ghz = Haswell @ 3ghz is wrong', because the 5960x turbos to 3.5ghz (so it would be 4 vs 3.5). The only thing that is strange about that though is the huge gap between Ivy and Haswell in this test, is this an AVX 2 enabled benchmark? Haswell didn't really have much ipc over Ivy in most test from what I remember...

It is not single thread. The huge gap between Haswell and Ivy in Dolphin is normal

@cdrkf

Its a synthetic benchmark thats why you cant really take it seriously. Probably most of what he has postet are synthetic ones.

It is not synthetic but a well-known emulator used by many people. I believe that jdwii uses it.

 

[8350rocks]

What about the 250ms hang in frame times from the 7700K that his graph showed? The 5960X also experienced a hang that was 180ms.

Considering that Ryzen never broke 30ms gap, and at no point was considered stuttering, I would think the hard numbers from the frame time graph would corroborate the user experience. Especially true if we are considering it was duplicated by multiple individuals playing the same games across the same systems....

There is your definitive proof that Ryzen was smoother than Intel. Frame times do not lie, and a 180-250ms hang in frame times would be a horrendously bad stutter with a noticeable effect on performance.

I'm going to point out that normal tech review sites have found absolutely no evidence of those types of hangs, especially when you consider that a 250ms hang time represents 16 frames being consecutively dropped. Which would be quite noticeable, even without latency benchmarking.

So I'll come out and say it: the test is being rigged is some way to make Intel appear worse. Because if the 7700k (or other, slower, Intel CPUs such as my 2600k) had those types of issues in normal usage, someone would have noticed by now. Or are you going to accuse every single tech site in existence of being Intel shills for not reporting this issue?

Gamer, he has the fraps capture from the hang, and it was run during the video.

Seriously...the test is not being rigged. He even comments that he would have thought Nvidia's drivers should have just dropped the single frame and moved on, it makes the most sense from his perspective. The drivers, however, did not do that. He checks signal interrupts and all sorts of things. If you develop on windows, he literally had EVERYTHING perfmon checks enabled to see what was going on the whole time.

I note FRAPS is very CPU intensive. It's a useful tool for getting FPS out of the GPU, but you are going to have latency hits, especially if Windows stuffs it on a Hyperthreaded core. i mean seriously, how many posts do we have daily about people complaining about shoddy gameplay while FRAPS is up and running? If you have a heavy CPU hog running in the background, CPU performance suffers. And by virtue of having more cores, Ryzen handles this case better.

Nothing to see here.

Why did the OC'ed 5960X also hang?

My guess? A major thread is getting assigned to a HTT core, and stalling out. Hence the downside to adding more and more threads: A greater and greater chance of a major program thread getting assigned to a SMT core and stalling out. I could confirm this if I had the raw perfmon logs, but based on observation and known Intel CPU performance, this explanation makes the most sense.

My point here is the test itself is invalid: You can't try and gauge CPU performance within an application while at the same time having another CPU intensive program running in the background.

That is just it, Wendell was monitoring EVERYTHING in perfmon, and could not come up with a good answer as to why.

 

[8350rocks]

https://forums.dolphin-emu.org/Thread-ryzen-1700x-stock-vs-dolphin-benchmark

Not to bad 10% lower then a 4770K meaning in this case Ryzen is below Haswell IPC but not by a massive margin

Better benchmark
http://www.legitreviews.com/wp-content/uploads/2017/03/dolphin-1700.jpg

I would ignore the first, because it is from one of those fake leaks released before launch. The performance reported in that leak wasn't confirmed by reviews.

And I would ignore the second because it is internally inconsistent. The 1700 and the 1800X are the same chip except for clocks, and the graph says that the 1700 OC to 4GHZ on all cores is slower that the 1800X at stock with 3.6GHz base and single core 4GHz turbo. A possible explanation of the discrepancy is that the 1800X is running with XFR activated in whose case the effective clocks aren't the clocks reported in the graph.

Here you have Dolphin benches from computerbase

Zen @3.6GHz performs like Haswell @3GHz in this bench.

The 6950X is right there as well...So it is accurate to say that Zen @ 3.7 GHz performs like Broadwell-E running turbo core up to who knows what.

 

[jdwii]

@cdrkf

Its a synthetic benchmark thats why you cant really take it seriously. Probably most of what he has postet are synthetic ones.

Actually that is a test from a gaming emulator dolphin its related to real performance numbers when using the program i know i use it all the time.

 

[juanrga]

Here you have Dolphin benches from computerbase

Zen @3.6GHz performs like Haswell @3GHz in this bench.

The 6950X is right there as well...So it is accurate to say that Zen @ 3.7 GHz performs like Broadwell-E running turbo core up to who knows what.

The 6950X has a turbo is lower than the base clock of the 1800X.

Zen @3.6GHz ~ Haswell-E @3GHz ~ Broadwell-E @3GHz

 

[Gon Freecss]

The 6950X has a turbo is lower than the base clock of the 1800X.

Zen @3.6GHz ~ Haswell-E @3GHz ~ Broadwell-E @3GHz

Haswell-E and Broadwell-E can clock higher than 3.5GHz through Turbo Boost 2.0 and 3.0 respectively.

 

[juanrga]

The 6950X has a turbo is lower than the base clock of the 1800X.

Zen @3.6GHz ~ Haswell-E @3GHz ~ Broadwell-E @3GHz

Haswell-E and Broadwell-E can clock higher than 3.5GHz through Turbo Boost 2.0 and 3.0 respectively.

Turbo 3 is not being used here and the reason why the 6850K beats the 6900K and the 6800K beats the 6950X.

 

[jdwii]

Also that benchmark looks extremely single threaded, so I imagine if there were any i3s on there they'd be just under the i7s and still above Ryzen.

Many people who are on dolphin fourms use the unlocked Pentium dual core since its actually really good once its overclock to 4.5Ghz or more with dolphin.

 

[Gon Freecss]

The 6950X has a turbo is lower than the base clock of the 1800X.

Zen @3.6GHz ~ Haswell-E @3GHz ~ Broadwell-E @3GHz

Haswell-E and Broadwell-E can clock higher than 3.5GHz through Turbo Boost 2.0 and 3.0 respectively.

Turbo 3 is not being used here and the reason why the 6850K beats the 6900K and the 6800K beats the 6950X.

Oh, alright.

 

[-Fran-]

I'll leave this here: https://www.dropbox.com/home/public?preview=computer_de_dolphin_bench.png

And: https://dolphin-emu.org/blog/2016/06/24/dolphin-50-release/

Cheers!

 

[jaymc]

Amazon has knocked $30 off the sticker price of AMD's Ryzen 7 1800X processor.

http://www.pcworld.com/article/3190636/components-processors/amazon-has-knocked-30-off-the-sticker-price-of-amds-ryzen-7-1800x-processor.html

 

[gamerk316]

That is just it, Wendell was monitoring EVERYTHING in perfmon, and could not come up with a good answer as to why.

I'd recommend using something akin to GPUView (which essentially just visualizes perfmon data) to figure out what the threads are doing then. One of them's getting interrupted, and I'd wager FRAPS is the culprit. I'd expect an i7 to power through though, unless somehow one of the major threads is getting put on a HTT core.

In any case, FRAPS is known to cause performance issues. Run the same test again, minus FRAPS, and I'd wager the problem goes away.

In regards to Dolphin, it doesn't scale beyond two cores, and it makes use of all CPU extensions, up to and including AVX2. So I'd expect never CPUs to show absurdly large performance boosts even if IPC doesn't move much (Such as the Sandy to Haswell jump).

 

[8350rocks]

The 6950X has a turbo is lower than the base clock of the 1800X.

Zen @3.6GHz ~ Haswell-E @3GHz ~ Broadwell-E @3GHz

Haswell-E and Broadwell-E can clock higher than 3.5GHz through Turbo Boost 2.0 and 3.0 respectively.

Turbo 3 is not being used here and the reason why the 6850K beats the 6900K and the 6800K beats the 6950X.

Why would they disable only one aspect of turbo core? Please link where they clearly state they disabled any of the turbo aspects of any chip.

 

[jdwii]

I'll leave this here: https://www.dropbox.com/home/public?preview=computer_de_dolphin_bench.png

And: https://dolphin-emu.org/blog/2016/06/24/dolphin-50-release/

Cheers!

I can't see the first link but yeah the second one the team is doing amazing things the fact that even a FX 8350 at 4Ghz can run dolphin fine still doesn't mean we can't tell the difference between IPC using it,

Ryzen is a different architecture at times when i compared the 1500X to 7700K i saw more IPC with ryzen but this was just as often as when i saw way less IPC(like 20-40%). On average who's to say Amd claims 10% within kabylake which is basically Haswell i say it depends, on average probably around Haswell maybe a little less.

I said it a LONG time ago but even if Ryzen got Sandy-ivy level IPC i'd call it a success that stands true today.

I've been thinking about selling my CPU+board+Ram to get ryzen just for benchmarking and testing as i love doing that(had a lot of fun testing CMT on my old 8350), still haven't decided 100% yet. Now a days even a 70$ processor could do what i do on my PC, i guess until the next big game comes out that i like the last one runs perfect on all modern CPUs(Resident Evil 7)

I do however really love simulation and open world games and those are CPU heavy and often (IPC+Frequency) dependent