Intel 270K Plus vs AMD 9950X

[MaxC]

Can anyone speak to the real term performance difference between these two chips for DAW work? My research suggests the Intel performs better at low latency but the AMD performs better at moderate to high latency. I'm more concerned with raw power for large multitrack mixes and FX chains than I am with live recording, so by that metric I lean toward the AMD.

However, I'd also like to limit the TDP to keep the case cool and quiet. Gemini estimates that CPU temps for the AMD running at 105W Eco Mode are comparable to the Intel running at 180W TDP. Supposedly the two chips will perform comparably when running at moderate latency under these power limits. In that case, it would seem to make more sense to save ~$200 and go with the Intel.

Does this comparison ring true? Are there variables I am not considering? I've not built a DAW in over a decade, so I'm hardly up to speed on the latest tech. Any advice would be appreciated.

[Pictus]

To have an idea about DAW performance
viewtopic.php?t=574244&start=75

To have an idea about TDP
https://www.techpowerup.com/review/inte ... us/23.html

Considering the price/performance, the Intel 270K looks more interesting.
Just limit the MAX CPU power and it will be silent.
Whatever your choice may be, check
viewtopic.php?t=621326
Learn how to tweak and undervolt the 270K. (Not overclock)


Intel cooler - Scythe Mugen 6 Dual Fan
https://pcpartpicker.com/product/rV9wrH ... mg-6000dbe


AMD Cooler - Thermalright Peerless Assassin 140
https://pcpartpicker.com/product/w8n9TW ... -140-black
Both good and not expensive.
https://hwbusters.com/best_picks/best-c ... e-busters/

For PSU, check
https://hwbusters.com/best_picks/best-a ... busters/9/

A PC case that is silent and not expensive
https://pcpartpicker.com/product/zysV3C ... e-lan207rx


BTW, Noctua interesting stuff

[MaxC]

Thank you for all of the links and info. Looking at Scan Pro Audio's DAWBench results, it would seem that the 270K Plus is not only more cost-efficient, but flat out outperforms the 9950X in most metrics. That surprised me and now has me leaning toward the Intel.

If anyone out there has practical experience with both of these chips and would care to comment, that would be awesome.

[BJ]

I got a new PC last week and I had them put the Intel Core Ultra 7 24 Core 270K Plus in it and so far it has been great you can run so many tracks with many plugins without a hiccup compared to my old PC. I'm still installing everything and playing with it testing stuff. I don't have any experience with AMD so I can't compare.

[Jim Roseberry]

Considering the price/performance, the Intel 270K looks more interesting.
Just limit the MAX CPU power and it will be silent.

Wish Intel had also released a refreshed 285k.

If you choose the right cooler, the 270k will be plenty quiet (while working with audio).
If you're doing 3D rendering or video rendering, fan RPMs are going to ramp-up because CPU cores will be maxed at/near 100%.
You can avoid this (somewhat) by limiting the CPU's max performance.
IMO, That kind of negates the point in having a higher-end CPU.
At that point, save some money and just buy a lower-tier CPU.

I've never seen a client try to use a condenser mic in close proximity while rendering video.
With DAW software, glitches will occur long before reaching 100% load across all cores.

[knob_monster]

the intel core ultra 7 270k plus has better price-to-performance and outperforms the amd 9950x in dawbench audio tests.

[MaxC]

the intel core ultra 7 270k plus has better price-to-performance and outperforms the amd 9950x in dawbench audio tests.

I ended up opting for the 270k Plus for those very reasons. Haven't built it yet, but I figured I couldn't go wrong for the price. Thanks, all.

[planet_b]

the intel core ultra 7 270k plus has better price-to-performance and outperforms the amd 9950x in dawbench audio tests.

I ended up opting for the 270k Plus for those very reasons. Haven't built it yet, but I figured I couldn't go wrong for the price. Thanks, all.

Which DAW are you going to run on it? In case Live 12, I'd be interested in some simple benchmarks.

[MaxC]

the intel core ultra 7 270k plus has better price-to-performance and outperforms the amd 9950x in dawbench audio tests.

I ended up opting for the 270k Plus for those very reasons. Haven't built it yet, but I figured I couldn't go wrong for the price. Thanks, all.

Which DAW are you going to run on it? In case Live 12, I'd be interested in some simple benchmarks.

I'm a long-time Tracktion user, so I was assuming I'd upgrade to Waveform. With that said, the Reaper results in DAWBench (featured in another thread) have me reconsidering. That particular benchmark showed Reaper handling 2-3x more plugins that Cubase and Studio One on the same rig .

[VitaminD]

That's the issue with DAW benchmarks -- the performance/results really depend on the DAW you use. So the numbers may look great for one CPU in Reaper but may not be as impressive in another (compared to a different CPU) such as Cubase.

It would be a huge undertaking but I think it would be useful to have per DAW benchmarks and then a larger body of CPUs tested. So 3000, 5000, 7000, and 9000 AMD all in each. And 10000, 11000, 12000, etc on up for Intel.

Most people hold on to a PC for awhile, so seeing not only what newest Intel vs AMD perform is useful. But also older gen to newer gen to get an idea of performance improvements vs cost.

[legendCNCD]

@VitaminD Isn't this kinda what Dawbench and such are doing with their measurements?

[VitaminD]

No it doesn't seem to me they are.

Their measurements are Reaper only, tested in different Reaper versions (across charts), and the systems aren't the same ones for each test. So results for 14th gen Intel aren't comparable to those done for 9th or 10th gen. Same goes for AMD side.

Each chart is only comparable between the results on that chart. For the results to be comparable across multiple charts, they'd have to use the same everything (Daw, Daw version, plugin versions, audio interface, drivers, systems/OS) between each. But the tests aren't that scientific.

So it is just a very broad result that can be gleaned across cpu generations. When review sites test videocards, for example, they try the same video cards in the same system, using the same software and software versions, driver versions, etc. And if there are differences in drivers, those are usually listed in the results. That way there are little to no elements that can spoil the results.

The controls on Dawbench aren't tight enough and the tests themselves aren't broad enough across different DAWs to accommodate how a CPU will perform in a different environment.

I pretty much only look at DAWbench results per chart as a result. Whatever is on the chart is generally comparable to whatever else is on that specific chart. But unfortunately I don't think the results can carry to previous charts. And that's the part I'd like to see. So one could say 'ok this CPU2019 performed X results and this CPU2026 performed Y results' since they were tested each in the same environment.

The next step would be a database (similar to Geekbench or Cinebench) where users can run an automated test for their DAW of choice, upload results, and then search results. That way results can be averaged by hardware and software. That would be a grand (but useful undertaking!). Sadly I don't think the pro audio scene is really that big to be interested in such an endeavor.