AMD Ryzen - Our first look and thoughts for audio.

[DJ Warmonger]

The CCX connects to the memory controller - which does run on RAM frequency (intel runs it near core frequency).

That's interesting, could the RAM frequency affect final performance? What was OP test run at?

[PurpleSunray]

The CCX connects to the memory controller - which does run on RAM frequency (intel runs it near core frequency).

That's interesting, could the RAM frequency affect final performance? What was OP test run at?

https://www.golem.de/news/ryzen-7-1800x ... 25996.html

It's an extensive test with lots of (low-level) details, in german only unfortunately.
They tests on various systems, with various UEFI tweaks.

No RAM type mentioned on the audio test blog.

[PurpleSunray]

- double post -

[pekbro]

Great article, seems another instance of AMD cutting corners and dropping the ball.
Much as I had guessed in the other thread.

Hopefully, the Ryzen will improve, or at least serve to motivate intel towards
more cores for the mainstream.

-Cheers

[nIGhT-SoN]

For those that want to read more about Ryzen, there is an AMA on reddit with AMD's President and CEO- Lisa Su, CPU Technical Marketing - Robert Hallock, CPU Business Development - James Prior.

https://www.reddit.com/r/Amd/comments/5 ... and_other/

LE: They are already working on Ryzen 2 and Ryzen 3, at least that's what I got from the link above.

[pljones]

I just want lots of fast RAM

[PurpleSunray]

btw. the articel nails down the problem exactly:

So the big problem here appears to be inefficiency at lower buffer rates. The ASIO buffer is throwing data at the CPU in quicker bursts the lower you go with the setting, so with the audio crackling and breaking up it seems that the CPU just isn’t clearing the buffer quickly enough once it gets to around 70% CPU load at those lower 64 & 128 buffer settings

Intel at this buffer setting looks to be hitting 85% or higher, so whilst the AMD chip may have more RAW performance to hand, the responsiveness of the rest of the architecture appears to be letting it down.

Lower buffer rates means more busy memory controller. He cannot load big a chunk of data to L3 and then do something else, but thereare lot so small chunks. What you see at the CPU load is exactly that bottleneck.
CPU (pure computing power) is about the same, but AMD memory controller can only load to it 70% then crackling starts. Intel controller throughput is higher, so it can drive the number-cruncher-until to 85% until pipeline latency starts to hit the human ear.

[DJ Warmonger]

btw. the articel nails down the problem exactly:

So the big problem here appears to be inefficiency at lower buffer rates. The ASIO buffer is throwing data at the CPU in quicker bursts the lower you go with the setting, so with the audio crackling and breaking up it seems that the CPU just isn’t clearing the buffer quickly enough once it gets to around 70% CPU load at those lower 64 & 128 buffer settings
Intel at this buffer setting looks to be hitting 85% or higher, so whilst the AMD chip may have more RAW performance to hand, the responsiveness of the rest of the architecture appears to be letting it down.

Lower buffer rates means more busy memory controller. He cannot load big a chunk of data to L3 and then do something else, but thereare lot so small chunks. What you see at the CPU load is exactly that bottleneck.
CPU (pure computing power) is about the same, but AMD memory controller can only load to it 70% then crackling starts. Intel controller throughput is higher, so it can drive the number-cruncher-until to 85% until pipeline latency starts to hit the human ear.

My current buffer setting is 512 samples. Never ever considered lowering it to 64 and didn't hear anybody actually does that. So I feel safe here, anyway

512 samples at 44100 is 11 milisecond delay, so who cares?

[PurpleSunray]

512 samples at 44100 is 11 milisecond delay, so who cares?

Well.. I don't. My 5820K is pretty much idle on my projects. Don't need anything new. I'm just a tech freak that loves to read about new stuff, lol

[fluffy_little_something]

The AMD stock is already falling due to disappointment with ryzen performance...
http://www.cnbc.com/2017/03/02/shares-o ... ance-.html

[Numanoid]

The AMD stock is already falling due to disappointment with ryzen performance...
http://www.cnbc.com/2017/03/02/shares-o ... ance-.html

But you are the guy who isn't interested in becoming rich as I remember.

[fluffy_little_something]

Indeed. But I read the headline and thought it was interesting. It has nothing to do with me. I don't own any shares of any company...

I hope AMD will manage to fix things quickly before ryzen gets a stain in its reputation from the very start.

[Numanoid]

Stock prices can fluctuate but they who like AMD will not abandon them nevertheless, especially at this early stage of the launch of a new product

[Aleatoriac]

Thanks for sharing those assessments.

I end up "upgrading" roughly every 4 years. Always been AMD up until 2015. We'll see in early 2019, probably. Maybe. If I have money and still exist in any meaningful way.

Oh, and buy my music!!! It'll give you feels! (Or it won't! But something will happen!) link in the signature! Got to sell sell sell!

[incubus]

512 samples at 44100 is 11 milisecond delay, so who cares?

Well.. I don't. My 5820K is pretty much idle on my projects. Don't need anything new. I'm just a tech freak that loves to read about new stuff, lol

I'd love to have one of those, BTW. 6 cores, OC'ed, I'm sure it's very nice. I cheaped out and simple bought a new mobo for what I had (the other was terribly defective) so out roughly 230 bones.

But the intel 6 core stuff has been highly effective according to virtually anyone with them.