Menu
What's new
By:
Filters

DAWbench

Pictus

Pictus

Senior Member


​

Ryzen 7000 Series X3D Performance Testing​












The new CPUs are factory overclocked, by tweaking we can reduce the
max wattage and use air cooler.
Intel Core i9-13900K vs. AMD Ryzen 9 7950X at 125W and 65W
www.club386.com

Intel Core i9-13900K vs. AMD Ryzen 9 7950X at 125W and 65W | Club386

Prefer to keep temperature and power consumption down to lower levels this winter? Here's what happens when the best CPUs are scaled back.
www.club386.com www.club386.com









DAWbench 2021 Suite - Intel 12th Gen Results.
gearspace.com

Gearspace.com - View Single Post - DAWBench DSP / VI Universal - Cross Platform DAW Benchmarks :

Post 15749620 -Forum for professional and amateur recording engineers to share techniques and advice.
gearspace.com

Oq9ic7d.jpg

DQgoDCx.jpg

NsDx5Pl.jpg

F1VclwP.jpg




Click to read more...
 
Last edited:
So, judging from the top two charts, to get maximim performance I need an AMD system to run my FX plugins and an Intel system to run my VIs (i.e., neither system is better at both)? :sad:
 
Anthony said:
So, judging from the top two charts, to get maximim performance I need an AMD system to run my FX plugins and an Intel system to run my VIs (i.e., neither system is better at both)? :sad:
Click to expand...
The new AMD chips with 3D cache will change this soon or diminish the difference.
www.guru3d.com

AMD EPYC-CPUs based on 3D V-Cache and teases Zen 4 as well as 128-core Bergamo

AMD announced several new EPYC processors during a virtual event. The 'Milan-X' processor is a variation of the currently available Zen 3-based CPUs with 3D V-Cache. The proc will become available i...
www.guru3d.com www.guru3d.com
 
The Passmark results are interesting for CPUs in the $600-$700 range... it appears at first that the Ryzen 5950 smokes the Intel i9-12900K with the overall scores of 46K for AMD and 37K for Intel, and that's probably true for gamers. But for us audio maniacs, we need single-core and FPU performance and the new Intel smokes the Ryzen significantly in single-core shown on this chart, and then I saw the FPU results and the Intel was about 20% faster there as well. For MIDI work the Intel is cheaper and the better choice for under $750:


This chart shows a higher "operating cost" but if your purpose is to save money in that realm, not sure if this is the world for you, lol!

As far as the new AMDs with all the cache improvements, let's see the specs once they are out there. And the price... only the 16/32 processor will probably be under $1K. But WOW imagine having 128 cores?? Nutty. And we'll need the DAW companies to step up to leverage that of course. Bottom line, it just keeps getting better. After 11 years of owning almost every iPad model, the new M1 with StaffPad is just plain amazing.
 
damn I was hoping DDR5 didnt show such a big difference so that I could feel comfortable opting for a DDR4 motherboard and cheaper DDR4 ram hahaha sadly it is pretty much impossible to get DDR5 right now, really hoping early 2022 we will see more availability, been wanting to make the switch back to PC for a while now.
 
thevisi0nary said:
Good showing though I wish bussing was more involved in dawbench
Click to expand...
I am in the process of developing a bussing extended version of the DSP test.

I will be adding multiple group busses with assigned resource heavy pre-loaded tracks, and then having a secondary load metric for vertical ( serial) processing.

Still fleshing out the finer details and logistics , coming in 2022.
rgames said:
Thank goodness I can finally add that 400th compressor I've been needing :)
Click to expand...
Right, you continue to miss the forest for the tree's :)

Disclaimer for those not as well versed as Richard on all things DAW benchmarking.

DAWbench is a parallel (horizontal ) multiprocessing benchmark , it measures x-scaling ( x being the number of cores ) in a DAW environment at respective latency settings, the plugins or the voices are simply the metric to apply the incremental load.

Your mileage may vary how that information translates to respective working environments, session logistics and configuration.
SyMTiK said:
damn I was hoping DDR5 didnt show such a big difference so that I could feel comfortable opting for a DDR4 motherboard and cheaper DDR4 ram hahaha sadly it is pretty much impossible to get DDR5 right now, really hoping early 2022 we will see more availability, been wanting to make the switch back to PC for a while now.
Click to expand...
The DDR5 combined with the new dual memory controller array on the Z690 has definitely shone a light on memory bandwidths importance for the heavier Kontakt based sessions, and explains why the X299 has remained so strong in those session environments.

www.igorslab.de

DDR5 Deep Dive - Exclusive interview with Kingston about the new memory standard and many examples from practice | Page 2 | igor´sLAB

The basic changes to the new DDR5 memory standard have already been discussed around the launch of the Intel Alder Lake CPUs, even if sometimes not completely correct. So today we’re going to revisit…
www.igorslab.de www.igorslab.de

This is a good article explaining the Z690's memory sub system, and details the advantages of even the DDR4 array over the previous single memory controller architecture, and why the DDR5 configuration is another level above.

Its a shame Intel have essentially done another paper launch , not that AMD are any better btw. DDR5 is near non existent, and what is available is very limited and very expensive. I am yet to see many memory options for 64GB , let alone for 128GB , and the costing has to take a deep dive before it seriously becomes viable. ATM its a rort !

Peace
 
Last edited:
TAFKAT said:
I am in the process of developing a bussing extended version of the DSP test.

I will be adding multiple group busses with assigned resource heavy pre-loaded tracks, and then having a secondary load metric for vertical ( serial) processing.

Still fleshing out the finer details and logistics , coming in 2022.
Click to expand...
Sweet
 
Well this is depressing because I just bought a i9 12900k with a DDR4 motherboard because there isn't any DDR5 32GB Ram modules out there yet. And if now if I upgrade in a few months I will have to reinstall windows, software, pluging.....the works :emoji_expressionless:
 
Jeffrey Peterson said:
Well this is depressing because I just bought a i9 12900k with a DDR4 motherboard because there isn't any DDR5 32GB Ram modules out there yet. And if now if I upgrade in a few months I will have to reinstall windows, software, pluging.....the works :emoji_expressionless:
Click to expand...
Why do you have to do that?

I have more than once successfully ported whole Win10 instances from 10 year old legacy BIOS-based PCs to the most recent stuff with UEFI. Much faster than a reinstall even considering the 2 hours of research a few years ago!
My DAW was initially installed on an i7 8700 an successfully works till today on an 11700 after a mainboard crash.

So a simple board change on a recently new installed system to a relatively similar platform does surely NOT need a reinstall. At first start there is a short hardware scan, after the a restart you log in and install the newest drivers for the platform. That's it.
 
widescreen said:
Why do you have to do that?

I have more than once successfully ported whole Win10 instances from 10 year old legacy BIOS-based PCs to the most recent stuff with UEFI. Much faster than a reinstall even considering the 2 hours of research a few years ago!
My DAW was initially installed on an i7 8700 an successfully works till today on an 11700 after a mainboard crash.

So a simple board change on a recently new installed system to a relatively similar platform does surely NOT need a reinstall. At first start there is a short hardware scan, after the a restart you log in and install the newest drivers for the platform. That's it.
Click to expand...
mhmm thats encouraging. I think I just tried it once and it didn't work well. But I may try it!
 
You must log in or register to reply here.
Top Bottom