AMD new CPU RYZEN using blender to render.

I recommend to leave discussions of ES xeons for another thread. This thread is about the Ryzen CPU and Blender benchmark.

Otherwise we’ll have various topics in this tread (eg. AMD sponsoring blender conference - all videos having that text :P)

Now the RyZen CPU. I do wonder, if they tried to optimize the tile setting? 32x32 tiles, sure means some cache on each core.

My Xeon e5-2687w performs best at 16x16. 32x32 gives me 26s, where as 16x16 just under 25s.

Anyone else had similar experience by dropping the tile size?

All of them are Intel’s property, you’re not allowed to sell them.

Since the amount is small and limited I think Intel doesn’t want to stop this. Many traders (mostly private) are doing this for years openly on Ebay.
As an average private user like me, you normally don’t have much usage for them.
Business users won’t use them for many reasons (efficiency, quality, legality, warranty …)

Anyone else had similar experience by dropping the tile size?

I always render with the smallest possible tile size, which in my case is 8x8.
But this thumb rule applies to many-core CPUs only.

Using 56 threads (28 cores)

32x32: 00:10.35
8x8: 00:09.60

I’ll double check on mine. When I ran it at various tiles 16x16 gave me best results. though I guess if you have more tiles, there are less cores idle towards end of render. Will test it out on a larger scene.

Still 56 threads (28 cores)… drool.

Should probably kick off a Multithread CPU thread for any additional detailed discussions on that.

Still do wonder if they could have gotten more out of the RyZen if they adjusted the tiles and other parts of the scene. aprox 36s for 8 core puts it on par with 8 core Intel chips. This is good news as it could mean Intel will adjust their prizes to counter AMD release.

End result, costumers benefit.

I guess with few more weeks to CES, we should know more.

i7 4790K stock (4GHz) renders in 43.36 regardless of tile size (16 or 32) on linux.
<strike> So far RYZEN doesn’t look too impressive an improvement, 2 years later. </strike>

Rebooting to W10 to check.

PS On the ppa build it goes down to 42.67

EDIT: Holy crap, I suspected Linux was probably faster but 58.52? Wtf are they doing to the poor CPU?

I’d just like to interject for moment. What you’re refering to as Linux, is in fact, GNU/Linux, or as I’ve recently taken to calling it, GNU plus Linux. Linux is not an operating system unto itself, but rather another free component of a fully functioning GNU system made useful by the GNU corelibs, shell utilities and vital system components comprising a full OS as defined by POSIX.

Many computer users run a modified version of the GNU system every day, without realizing it. Through a peculiar turn of events, the version of GNU which is widely used today is often called Linux, and many of its users are not aware that it is basically the GNU system, developed by the GNU Project.

There really is a Linux, and these people are using it, but it is just a part of the system they use. Linux is the kernel: the program in the system that allocates the machine’s resources to the other programs that you run. The kernel is an essential part of an operating system, but useless by itself; it can only function in the context of a complete operating system. Linux is normally used in combination with the GNU operating system: the whole system is basically GNU with Linux added, or GNU/Linux. All the so-called Linux distributions are really distributions of GNU/Linux!

Wtf are they doing to the poor CPU?

The GNU Compiler Collection generates faster code than the Microsoft compiler.

I know, I know. I’m just using the colloquial term. I now wonder if a gcc build for windows would be faster (I suppose the official build is MSVC?)

MingW builds (which would be GCC for Windows) is faster than those compiled with Visual Studio, but it’s not a priority since they were previously known to have stability issues.

The larger priority right now is to simply have compatibility with the latest versions of VS.

:nod:

Good info to know. Makes reading benchmark results much harder than I thought.

Could it be a matter of optimization flags not used by default with MSVC?

That is a significant difference. hmm… So using horrible math, my 26s in windows would be around 18 in Linux/GNU/etc…

So that also brings the question, when AMD did benchmarks with their CPU, was it Windows or Linux? Anyone recalls?

I’ll so have to test the GCC built and then eventually Linux…

I always render with the smallest possible tile size, which in my case is 8x8.But this thumb rule applies to many-core CPUs only.

Using 56 threads (28 cores)

32x32: 00:10.35
8x8: 00:09.60

Same machine running windows 7 using Blender Daily Build:
Tile: 8x8
Render time first run: 00.12.76
Second run: 00.11.75


And Blender 2.78a
Tile: 8x8
Render time first run: 00.17.28
Second run: 00.16.19

What is this? Intel thread? Comparison of fat, chocolate VS bacon?
Few good posts, all other spam. Keep it up :rolleyes:.

Intel thread?

Nope. What are CPUs made for?
I think there are enough people with limited wallets.

It’s all about how much do you get for a certain amount of money.

As AMD have allready precise it, and the press who have cover itt, all tests was conducted on Windows 10 x64 with the official Blender build … And this dont change anything as intel and AMD cpu’s was running on the same OS, system etc

Title Thread: AMD new CPU RYZEN using blender to render.

Sarcasm. :wink:
Tho still off… Price wise, all assumptions are still speculations… so. Nothing else to add.

Anyone knows what to look for & how it behaves on encoding tasks and with openCL, sims and such?
IMO, Rendering is soon to become irrelevant considering working hours.

Thanks Lane. That means I’m fine with what I got (for now).

I’m still keen to see what Ryzen can do at higher clocks, results were shown only with 3.4 (or there about) without any turbo. All in a nice 95w envelope… dreamy.

Now to see where it ends up on the price scale against the i7-7700k. Can’t wait for competition to restart again. In the end we all win. Now if it was as easy to improve your blender skills as just upgrading a CPU :stuck_out_tongue:

luckily CES is just around the corner, wonder if they are planning to show off more blender?

One thing I really like about this, is the free advertisement for Blender. More interest, hopefully will mean more funding, meaning better product for us all.

With regards to Intel concerns, we are just testing to see where we fit in against the yet not released product. :stuck_out_tongue: I’m afraid to boot up my FX-8350 to see where it stacks up.

The rumor has it that ryzen will over clock to 5ghz on air cooling…

That would be a nice thing to see assuming Ryzen is already matching or surpassing Intel’s pricier chips.

Unfortunately, it’s hard to take anything at face value right now with the rumor mill working as hard as it is.

that rumour was with just one core enabled … so not much of an achievement. :frowning: Still we’ll know soon enough. CES is just around the corner and AMD is to (supposedly) release RyZen during that time, or at least provide a lot more information on it