The No.1 Website for Pro Audio
Anyone interested in my new benchmark to replace the dated Evan's test?
Old 5th October 2019
  #151
Here for the gear
 

Thanks for the reply. It could be that older versions of Logic run more tracks. I have 40 gb of 2666 crucial ram, so probably this won’t be the bottleneck. Sorry to hear about yout Imac, but I would be cautious with ordering the 580x gpu, as it seems to be running hotter then the Vega 48 according to threads on macrumors. It could have an influence on the daw performance under stress.
Old 6th October 2019
  #152
Here for the gear
Quote:
Originally Posted by chezcharleso View Post
Thanks for the reply. It could be that older versions of Logic run more tracks. I have 40 gb of 2666 crucial ram, so probably this won’t be the bottleneck. Sorry to hear about yout Imac, but I would be cautious with ordering the 580x gpu, as it seems to be running hotter then the Vega 48 according to threads on macrumors. It could have an influence on the daw performance under stress.
I totally agree with your comment on the Vega 48. My next iMac i9, and probably Vega 48, I hope that will last more than the previous one. At least I will not play demanding games during summer heatwaves and I will spend more time making music in the studio with the iMac
Old 7th October 2019
  #153
Quote:
Originally Posted by Jim Rosebrook View Post
Have you re-tested your 2019 8-core MBP using latest Logic Pro 10.4.6?

And at 44.1 and 128 Buffer?

Really curious how it compares to my 2018 i9 MBP 6-core... I'm getting 76 tracks.
73 tracks here on the MacBook Pro 2018 I7 2.2. those 8 cores getting closes to 100tracks are beasts.
Old 4 weeks ago
  #154
Gear Nut
 
derdiedasscheusa's Avatar
 

110 Tracks for a 5.1 Mac Pro 2x 3.33Ghz at 44.1Khz.

Did anyone test the 13" 1.4Ghz MBP 2019 yet? I am very interested!
Old 4 weeks ago
  #155
Lives for gear
Quote:
Originally Posted by derdiedasscheusa View Post
110 Tracks for a 5.1 Mac Pro 2x 3.33Ghz at 44.1Khz.

Did anyone test the 13" 1.4Ghz MBP 2019 yet? I am very interested!
Think I remember it being in the low 90s. Should be a page or two back.
Old 4 weeks ago
  #156
Lives for gear
Hey all,
I've been culling the data from this thread in an attempt to correlate Geekbench-5 performance figures with the new TNM LPX test suite. The intent is to eventually develop a reliable way of predicting the audio performance figures of the TNM LPX test by knowing the Geekbench 5 figures (or vice versa).

I've attached the current rough state of the spreadsheet as part of my plea for help. For those of you like @ zedsdeadbaby and perhaps @ poshook it would be extremely helpful to everyone to run the Geekbench 5 tool on your machine in order to establish the effects of overclocking (if any) since the standard Geekbench figure wouldn't necessary apply.

It probably is a good time now to explain the MultiFactor and Linearity columns. (I know I need better names). MultiFactor simply divides the multi-core Geekbench 5 result by the single-core result. That sort of gives an idea of the net number of cores achieved by this particular multi-core chip. The 'Linearity' column simply divides the MultiFactor by the number of physical cores to get an idea of how close to optimal the implementation of this chip is compared to other chips.

This was prompted by @ poshook who has the high water mark of results among the group, and the Linearity figure stood out at an almost perfect 0.98. The only other CPUs in this collection listed that have such high Linearity numbers are the ancient x5670, x5680 chips, the i7-6700k, and a few others including the E5-1650 from the 6-core 2013 Mac Pro.

It's possible that these CPUs distinguished themselves as better-than-average performers in the subjective user experience for DAW work. This is one of the things I hope to learn by collecting more data examples and user experience commentary. Of course, there's no substitute for raw horsepower, and that's why the ancient 12-core 2009/2010/2012 Mac Pro machines continue to post very high track count results in the TNM tests despite their poor linearity.

Thanks in advance. Hopefully we'll all learn something useful.

Please excuse the finger-happy multiple attachments. They're identical.
Attached Thumbnails
Anyone interested in my new benchmark to replace the dated Evan's test?-cpu-vs-tnm-20191115a.jpg   Anyone interested in my new benchmark to replace the dated Evan's test?-cpu-tnm-20191115a.jpg   Anyone interested in my new benchmark to replace the dated Evan's test?-cpu-tnm-20191115a.jpg  
Old 4 weeks ago
  #157
Gear Nut
 
derdiedasscheusa's Avatar
 

MacBook Pro mid 2012 2,5 Ghz i5 13" 4GB RAM Sata SSD / Mojave 10.14.6, Logic 10.4.7:

12 tracks

I also attached my Geekbench 5 scores. Single core score 616 / Multi core score 1301.
Attached Thumbnails
Anyone interested in my new benchmark to replace the dated Evan's test?-macbook-pro-9-2-mid-2012-13-22-i5-4gb-.jpg  
Old 4 weeks ago
  #158
Gear Nut
 
derdiedasscheusa's Avatar
 

Quote:
Originally Posted by MediaGary View Post
Hey all,
I've been culling the data from this thread in an attempt to correlate Geekbench-5 performance figures with the new TNM LPX test suite. The intent is to eventually develop a reliable way of predicting the audio performance figures of the TNM LPX test by knowing the Geekbench 5 figures (or vice versa).

I've attached the current rough state of the spreadsheet as part of my plea for help. For those of you like @ zedsdeadbaby and perhaps @ poshook it would be extremely helpful to everyone to run the Geekbench 5 tool on your machine in order to establish the effects of overclocking (if any) since the standard Geekbench figure wouldn't necessary apply.

It probably is a good time now to explain the MultiFactor and Linearity columns. (I know I need better names). MultiFactor simply divides the multi-core Geekbench 5 result by the single-core result. That sort of gives an idea of the net number of cores achieved by this particular multi-core chip. The 'Linearity' column simply divides the MultiFactor by the number of physical cores to get an idea of how close to optimal the implementation of this chip is compared to other chips.

This was prompted by @ poshook who has the high water mark of results among the group, and the Linearity figure stood out at an almost perfect 0.98. The only other CPUs in this collection listed that have such high Linearity numbers are the ancient x5670, x5680 chips, the i7-6700k, and a few others including the E5-1650 from the 6-core 2013 Mac Pro.

It's possible that these CPUs distinguished themselves as better-than-average performers in the subjective user experience for DAW work. This is one of the things I hope to learn by collecting more data examples and user experience commentary. Of course, there's no substitute for raw horsepower, and that's why the ancient 12-core 2009/2010/2012 Mac Pro machines continue to post very high track count results in the TNM tests despite their poor linearity.

Thanks in advance. Hopefully we'll all learn something useful.

Please excuse the finger-happy multiple attachments. They're identical.
It would be very helpful if you could sort them by performance, high to low.
Old 4 weeks ago
  #159
Gear Nut
 
derdiedasscheusa's Avatar
 

Again: Mac Pro 5,1 2010 12 Core dual X5680 3.33Ghz Mojave 10.14.6, Logic 10.4.7, RX580 Pulse, 970 EVO NVME:


110 Tracks


Geekbench 5 single core 649 / multi core 6739.
Attached Thumbnails
Anyone interested in my new benchmark to replace the dated Evan's test?-macpro-5-1-2010-12-core-x5680.jpg  
Old 4 weeks ago
  #160
Lives for gear
Quote:
Originally Posted by derdiedasscheusa View Post
It would be very helpful if you could sort them by performance, high to low.
The tab of the spreadsheet that I posted is sorted high-to-low on the 3rd column which is the multi core Geekbench score. I have other tabs sorted by single core and by linearity.

When I get enough data rows of theTNM LPX Benchmarks I will absolutely post a tab sorted in that order too.

Thanks for helping!
Old 3 weeks ago
  #161
Lives for gear
Quote:
Originally Posted by MediaGary View Post
Hey all,
I've been culling the data from this thread in an attempt to correlate Geekbench-5 performance figures with the new TNM LPX test suite. The intent is to eventually develop a reliable way of predicting the audio performance figures of the TNM LPX test by knowing the Geekbench 5 figures (or vice versa).

I've attached the current rough state of the spreadsheet as part of my plea for help. For those of you like @ zedsdeadbaby and perhaps @ poshook it would be extremely helpful to everyone to run the Geekbench 5 tool on your machine in order to establish the effects of overclocking (if any) since the standard Geekbench figure wouldn't necessary apply.

It probably is a good time now to explain the MultiFactor and Linearity columns. (I know I need better names). MultiFactor simply divides the multi-core Geekbench 5 result by the single-core result. That sort of gives an idea of the net number of cores achieved by this particular multi-core chip. The 'Linearity' column simply divides the MultiFactor by the number of physical cores to get an idea of how close to optimal the implementation of this chip is compared to other chips.

This was prompted by @ poshook who has the high water mark of results among the group, and the Linearity figure stood out at an almost perfect 0.98. The only other CPUs in this collection listed that have such high Linearity numbers are the ancient x5670, x5680 chips, the i7-6700k, and a few others including the E5-1650 from the 6-core 2013 Mac Pro.

It's possible that these CPUs distinguished themselves as better-than-average performers in the subjective user experience for DAW work. This is one of the things I hope to learn by collecting more data examples and user experience commentary. Of course, there's no substitute for raw horsepower, and that's why the ancient 12-core 2009/2010/2012 Mac Pro machines continue to post very high track count results in the TNM tests despite their poor linearity.

Thanks in advance. Hopefully we'll all learn something useful.

Please excuse the finger-happy multiple attachments. They're identical.
Sure, I can run geekbench 5 on the machine. I'm in the middle of a project, but will post back ASAP...

When you mention overclocking - my machine isn't overclocked. Are you asking if I could do a geekbench of both non-overclock and overclocked?
Old 3 weeks ago
  #162
Lives for gear
Quote:
Originally Posted by zedsdeadbaby View Post
Sure, I can run geekbench 5 on the machine. I'm in the middle of a project, but will post back ASAP...

When you mention overclocking - my machine isn't overclocked. Are you asking if I could do a geekbench of both non-overclock and overclocked?

Something I read was interpreted as your machine not running at the stock clock speed. If it's at the stock speed then don't bother to run the Geekbench test.

I appreciate your willingness to help, but don't break your stride to accommodate something like this. It'll keep.

It's only the overclocked examples that need the additional documentation. The small percentage differences between stock machines shouldn't be enough to upset the quest for worthwhile insights.

Thanks again.
Old 2 weeks ago
  #163
Gear Nut
 
derdiedasscheusa's Avatar
 

New Hackintosh system I am testing now: i9 9900K, 64GB 3600Mhz RAM, RX 580 with 4K Display attached, Samsung NVME, this thing is a beast and it is sooo damn silent!

I really feel like the Hackintosh is just the perfect thing for me. After two failed attempts in the past, now it seems to work just fine!

It runs on stock frequencies, no overclocking has been done.

153 tracks

Geekbench 5: single core 1283 // multi core 9622
Attached Thumbnails
Anyone interested in my new benchmark to replace the dated Evan's test?-bildschirmfoto-2019-11-24-um-15.41.54.jpg  
Old 2 weeks ago
  #164
Lives for gear
Quote:
Originally Posted by derdiedasscheusa View Post
New Hackintosh system I am testing now: i9 9900K, 64GB 3600Mhz RAM, RX 580 with 4K Display attached, Samsung NVME, this thing is a beast and it is sooo damn silent! ...
Thanks for the data! This is an impressively strong performance and the "Linearity" metric that I've calculated is 0.94 which is 'way up there. The theory is holding together within this benchmark.

I have high hopes that @ Sascha Franck , who has some thoughts on building a another benchmark that stresses different aspects of soft-synth behavior, will bear fruit. We need a more "3D" approach to analyzing/predicting what makes a good DAW machine, and the combination of benchmarks may be the answer.

Thanks again.

EDIT: Here's the link to the comment from Sascha Franck [At last: new Mac Pro! ]

Last edited by MediaGary; 2 weeks ago at 02:57 PM.. Reason: added link to Sascha Franck insight
Old 2 weeks ago
  #165
Lives for gear
 

Quote:
Originally Posted by MediaGary View Post
Thanks for the data! This is an impressively strong performance and the "Linearity" metric that I've calculated is 0.94 which is 'way up there. The theory is holding together within this benchmark.

I have high hopes that @ Sascha Franck , who has some thoughts on building a another benchmark that stresses different aspects of soft-synth behavior, will bear fruit. We need a more "3D" approach to analyzing/predicting what makes a good DAW machine, and the combination of benchmarks may be the answer.

Thanks again.

EDIT: Here's the link to the comment from Sascha Franck [At last: new Mac Pro! ]
Sorry for being lazy. I'll try my best to come up with at least an idea soon-ish-ly.
Old 1 week ago
  #166
Lives for gear
 
timtoonz's Avatar
Quote:
Originally Posted by derdiedasscheusa View Post
New Hackintosh system I am testing now: i9 9900K, 64GB 3600Mhz RAM, RX 580 with 4K Display attached, Samsung NVME, this thing is a beast and it is sooo damn silent!

I really feel like the Hackintosh is just the perfect thing for me. After two failed attempts in the past, now it seems to work just fine!

It runs on stock frequencies, no overclocking has been done.

153 tracks

Geekbench 5: single core 1283 // multi core 9622
I'm looking at putting together a Hackintosh system. Is there a secret to getting a 'silent' machine? A special case or fan? The rest (CPU, ram, motherboard) seems relatively straight forward.
Old 1 week ago
  #167
Gear Nut
 
derdiedasscheusa's Avatar
 

Quote:
Originally Posted by timtoonz View Post
I'm looking at putting together a Hackintosh system. Is there a secret to getting a 'silent' machine? A special case or fan? The rest (CPU, ram, motherboard) seems relatively straight forward.
No real secret, just buy the most silent and/or cooling efficient components you can get. Many of them have noise levels mentioned so check the db‘s and go for the most silent ones.

My parts are: BeQuiet! silent base 600, Noctua U12S chromax, Corsair RM750x, Artcic MX4 paste. This thing is way more quiet than my Mac Pro. It also seems that the times when the fans ramp up its more like shorter bursts and it gets cooled imeadately. Whereas the Mac ramps up and tends to stay there for longer periods of time.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump