i7 Builds - Specs and Results

[Bob Olhsson]

I think they came out before M2 but are the same idea. The price was astronomical until recently.

[noiseboyuk]

Just found this thread (thanks guitardom) with a new rig on the way.

Dverb Tests for my current rig 4930k, RME Babyface @1024 PT 2018.3 - 82 tracks, or 410 instances.

[guitardom]

Originally Posted by noiseboyuk

Just found this thread (thanks guitardom) with a new rig on the way.

Tests for my current rig 4930k, RME Babyface @1024 - 82 tracks, or 410 instances.

Hey Guy, be sure to specify those are Dverb test numbers so everyone understands what you are talking about. Also specify which Pro Tools version. It also helps to do a test at 64 HW Buffer and 1024. The outline and a template is here

Your numbers seem a bit low though.

[noiseboyuk]

Good shout, amended the post. As I posted on the other thread, I did the high performance test @64 and managed zero tracks! Even one track overloaded the CPU. That is, surely, pretty poor....

The rig has actually been very solid for me and PT on 512, but it's not been performing well in Cubase, hence getting a new one.

[noiseboyuk]

Thinking a little more about my failed performance test (2nd test here - http://www.pro-tools-pc.com/performa...ing-pro-tools/ ) , perhaps its not so surprising? This test is pretty tough on single core performance, and what my result appears to show is that mine isn't at a level to run it all. If I could make one, I might well be able to make many more channels, its that it stumbles at the first hurdle as it were.

I mentioned that I have issues in Cubase and not PT, and this is largely because I don't use VIs much in PT (PT = audio post, Cubase = music composition in my world). My PT projects don't have any very demanding single channels, the load is well spread, whereas this is not the case in Cubase.

My Geekbench 4 score on the 4930k is 3500, I'd be curious to know if others on a similar score are unable to run the Performance Test 2. It's all a little moot in the sense that today my 7820x rig arrives, but knowledge is always a good thing...

[noiseboyuk]

Well, I really can't get a whole lot of sense out of Performance Test 2. On the new 7820x, I can't get a single channel out of it at 64 either. At 128 the 4930k can get 12 instances, the 7820X gets 16 - basically the number of cores / threads.

Not sure this is a very useful test.

For the Dverb test, it was the opposite problem @1024 - I maxed out the number of channels. 254x5 = 1270, compared to only 410 on the 4930k. However, there too I am suspicious. Watching the system usage was odd - when I got to 254 channels all cores and Total hit about 95%, but it would only last for a few seconds, then it would always drop to near zero, as if it were an absolute breeze. I don't know what's causing that supreme efficiency, I hadn't seen this behaviour on the 4930k.

All round, not at all sure what to make of any of it.

[guitardom]

Originally Posted by noiseboyuk

Well, I really can't get a whole lot of sense out of Performance Test 2. On the new 7820x, I can't get a single channel out of it at 64 either. At 128 the 4930k can get 12 instances, the 7820X gets 16 - basically the number of cores / threads.

Not sure this is a very useful test.

For the Dverb test, it was the opposite problem @1024 - I maxed out the number of channels. 254x5 = 1270, compared to only 410 on the 4930k. However, there too I am suspicious. Watching the system usage was odd - when I got to 254 channels all cores and Total hit about 95%, but it would only last for a few seconds, then it would always drop to near zero, as if it were an absolute breeze. I don't know what's causing that supreme efficiency, I hadn't seen this behaviour on the 4930k.

All round, not at all sure what to make of any of it.

You have something going on. I hit over 140 tracks with the P2 test on them.

The results you are seeing on the dverb test are the opposite of what you think. When it starts dumping down like that, you are past the threshold of where the systems wants to run. You typically need to drop it down the tracks a bit to find where it stabilizes.

What interface do you run?? You have a problem that went from your first system and now on your second system. Interface driver is the most likely culprit.

[noiseboyuk]

It's in the sig - RME Babyface, with the latest drivers too. I'd be surprised if it were this as RME are known as some of the best in the business, but I do follow your logic.

EDIT - Looking at the charts here - https://www.gearslutz.com/board/musi...data-base.html - the Babyface does pretty well at low latencies - for the RXC @64 it scores 123, with 149 being the very highest and 55 the lowest, so it should be able to run the test fine you'd think.

[mrpollack]

Originally Posted by jlumbreras88

So I JUST switched from Mac to Windows this week after 12 years. My build seems pretty solid. I haven't had any issues yet.

Asus ROG Maximus X Hero
i7 8700k delidded 5.0 OC
Noctua d15 cooler
32gb RAM 3200
960pro SSD for system and apps
850 evo for samples
850 evo for sessions
1tb HDD for storage
1080ti because I game but completely not necessary
Thunderbolt card for my Ensemble
Corsair 540 air case

Hi
After reading your post some time ago ,I just bult my new PT Native system around Asus ROG Maximus X Hero and have some noise issues with this card. I've searched on internet and found out that there is some problem with this card (asus forum) ,and people get some noise and crackling at the background. Have you experienced anything like this ?
Thanks
jacek

[YYR123]

I would RMA the board

Most top end mobo companies have really good warranties for their top end products