Author Topic: Corona 1.3 Benchmark  (Read 254378 times)

2021-11-02, 13:41:17
Reply #225

recep

  • Users
  • *
  • Posts: 1
    • View Profile
Lower than expected results on 12th gen 12900K. Probably incompatible with Windows 11 or Hybrid architecture. Can't share the result due to NDA.

2021-11-03, 12:49:42
Reply #226

Dionysios.TS

  • Active Users
  • **
  • Posts: 766
    • View Profile
    • Evolvia Imaging
Lower than expected results on 12th gen 12900K. Probably incompatible with Windows 11 or Hybrid architecture. Can't share the result due to NDA.

I don't think is a Win 11 problem.
From other bechmarks online, is obvious that in multithreading the 12900K is not one of the best CPUs out there. An AMD Ryzen 5950X is still faster and in single thread is just a bit lower that the 12900K at the end. Excuse me for my comment but I don't see anything revolutionary from this CPU now...

2021-11-04, 09:31:47
Reply #227

Ryuu

  • Former Corona Team Member
  • Active Users
  • **
  • Posts: 654
  • Michal
    • View Profile
Well, it's kinda hard to judge anything from that comment without knowing what the expected and the actual measured results are :)

2022-02-14, 20:30:23
Reply #228

Marco Onersi

  • Users
  • *
  • Posts: 3
    • View Profile
Good evening everyone,

I post my first benchmark with Corona Render.

I am a CGI enthusiast who has not been in the business for a few years.

I am getting closer again and I have seen the potential of this CPU software.

My current workstation:
Xeon E52697v2
48Gb Ram
Nvidia Gtx 970

it is several years old but it is still in perfect working order and I haven't used it for at least 5 years.

I would love to hear from you taking into account that I currently use 3ds Max 2015 and Vray 3.6 and a "decent" 1920 * 1080 image with about two and a half hours I can get it.

Probably the best test would be to buy a month of software and try.

Thanks for any advice.

Corona 1.3 Benchmark Finished
BTR Scene 16 passes
Intel (R) Xeon (R) CPU E5-2697 v2 @ 2.70GHz
  Real CPU Frequency [GHz]: 2.6
Render Time: 0:02:20, Rays / sec: 3.447.620

Greetings,
Marco

2022-02-15, 08:53:16
Reply #229

Ryuu

  • Former Corona Team Member
  • Active Users
  • **
  • Posts: 654
  • Michal
    • View Profile
Probably the best test would be to buy a month of software and try.

You don't need to buy Corona for this. We offer a 45-day demo license. All you need to do is just to download the installer from web, install Corona and then when you try to use Corona in 3ds Max, you'll be prompted to activate your license where you'll be given the choice to activate the demo.

2022-02-15, 11:58:59
Reply #230

Marco Onersi

  • Users
  • *
  • Posts: 3
    • View Profile
Thank's for your reply

2022-02-17, 12:55:45
Reply #231

Stan_But

  • Active Users
  • **
  • Posts: 526
    • View Profile
    • https://www.behance.net/archdizs
Hello everyone!

Cant run Corona renderer Benchmark with new clean OS win 7 x 64. Benchmark requests license every time. Does it need Corona's license for testing PC?

2022-05-31, 04:54:04
Reply #232

AVWS

  • Users
  • *
  • Posts: 1
    • View Profile
Corona 1.3 Benchmark Finished
BTR Scene 16 passes
AMD Ryzen Threadripper 2990WX 32-Core Processor (x4)
 Real CPU Frequency [GHz]: 3.3
Render Time: 0:00:43, Rays/sec: 11,285,800

....I think I have the best time around here.. I'm impressed.. and I was thinking of upgrading this PC to a Threadripper PRO 3975WX....
« Last Edit: 2022-05-31, 05:01:34 by AVWS »

2022-06-19, 17:10:17
Reply #233

fobus

  • Active Users
  • **
  • Posts: 388
    • View Profile
Dual EPYC 7T83 with 256 GB of old 2666 MHz RAM - 9 sec

2022-06-20, 13:58:15
Reply #234

alexyork

  • Active Users
  • **
  • Posts: 699
  • Partner at Recent Spaces
    • View Profile
    • RECENT SPACES
I know this has all been requested years ago but I'd really love to see an updated benchmark that has the following:

- ability to run multiple tests automatically and take an average (and best) result
- something that's at least taking 5 or 10 x longer than currently on an average machine, to allow proper time for any throttling to take real effect etc.
- something that uses multiple render elements with denoising (which stresses the CPU a lot more and helps judge temps)
- outputs the result to desktop automatically
Alex York
Partner
RECENT SPACES
recentspaces.com

2022-06-20, 14:41:01
Reply #235

fobus

  • Active Users
  • **
  • Posts: 388
    • View Profile
- something that's at least taking 5 or 10 x longer than currently on an average machine, to allow proper time for any throttling to take real effect etc.

No need to do this longer by x-times. It will be much better to run it for 1 or 2 minutes like in vray benchmark. This way it will be good for everyone.

2022-11-12, 20:55:46
Reply #236

artem456

  • Users
  • *
  • Posts: 1
    • View Profile
Corona 1.3 Benchmark Finished
BTR Scene 16 passes
11th Gen Intel(R) Core(TM) i7-11700F @ 2.50GHz
 Real CPU Frequency [GHz]: 4.4
Render Time: 0:01:32, Rays/sec: 5 243 820

2022-11-13, 20:02:01
Reply #237

Ondra

  • Administrator
  • Active Users
  • *****
  • Posts: 9048
  • Turning coffee to features since 2009
    • View Profile
I know this has all been requested years ago but I'd really love to see an updated benchmark that has the following:

- ability to run multiple tests automatically and take an average (and best) result
- something that's at least taking 5 or 10 x longer than currently on an average machine, to allow proper time for any throttling to take real effect etc.
- something that uses multiple render elements with denoising (which stresses the CPU a lot more and helps judge temps)
- outputs the result to desktop automatically
We are working on new benchmark to be released during v10 cycle, that will have everything mentioned here, probably except for the denoising (unless we add it as a separate benchmark, but it is still tricky, as denoising is a finite workload. We want the benchmark to run for a fixed amount of time this time)
Rendering is magic.How to get minidumps for crashed/frozen 3ds Max | Sorry for short replies, brief responses = more time to develop Corona ;)

2022-11-14, 12:40:02
Reply #238

alexyork

  • Active Users
  • **
  • Posts: 699
  • Partner at Recent Spaces
    • View Profile
    • RECENT SPACES
I know this has all been requested years ago but I'd really love to see an updated benchmark that has the following:

- ability to run multiple tests automatically and take an average (and best) result
- something that's at least taking 5 or 10 x longer than currently on an average machine, to allow proper time for any throttling to take real effect etc.
- something that uses multiple render elements with denoising (which stresses the CPU a lot more and helps judge temps)
- outputs the result to desktop automatically
We are working on new benchmark to be released during v10 cycle, that will have everything mentioned here, probably except for the denoising (unless we add it as a separate benchmark, but it is still tricky, as denoising is a finite workload. We want the benchmark to run for a fixed amount of time this time)

Awesome great to hear. I get the denoising issue. If you can please have it as a separate benchmark option that would be great - you could just prevent that one from submitting results to the website as they would skew the results. It's really more for internal testing/speed comparison/temps testing. Cheers!
Alex York
Partner
RECENT SPACES
recentspaces.com

2022-11-14, 17:20:11
Reply #239

burnin

  • Active Users
  • **
  • Posts: 1532
    • View Profile
IMO
for a longer, more intricate evaluation of a complex scene, instead of a still, an animation should do way better
in this regard, having options on start (to denoise and/or animate - select frames...) would be a must