Author Topic: Corona 1.3 Benchmark  (Read 254421 times)

2016-03-08, 20:14:08
Reply #30

PROH

  • Active Users
  • **
  • Posts: 1219
    • View Profile
In my testing of my home PC (i7-3930K) I've got five different result between 4;38 and 5;35. The "hang" after rendering differed from 4 sec. to over 20 sec.

2016-03-08, 20:25:17
Reply #31

sten

  • Active Users
  • **
  • Posts: 5
    • View Profile

2016-03-08, 20:40:24
Reply #32

PROH

  • Active Users
  • **
  • Posts: 1219
    • View Profile
My PC is warning me about this site.

2016-03-08, 22:30:23
Reply #33

SaY

  • Active Users
  • **
  • Posts: 72
    • View Profile
There is a long pause at the end of the render.
I'm getting inconsistent result with my E5-2699v3 - anywhere from 0:52 to 1:50  The rendering itself is finished at around 0:45.
does this happen also with regular Corona rendering?
No, I haven't noticed.
Also, looks like i'm not the only one:
In my testing of my home PC (i7-3930K) I've got five different result between 4;38 and 5;35. The "hang" after rendering differed from 4 sec. to over 20 sec.

2016-03-08, 22:42:21
Reply #34

Ondra

  • Administrator
  • Active Users
  • *****
  • Posts: 9048
  • Turning coffee to features since 2009
    • View Profile
we will tomorrow probably release updated version of the benchmark that reports consistent times in the popup window (the one sent to web seems OK), and runs even if MSVS fails to install
Rendering is magic.How to get minidumps for crashed/frozen 3ds Max | Sorry for short replies, brief responses = more time to develop Corona ;)

2016-03-09, 03:47:23
Reply #35

SaY

  • Active Users
  • **
  • Posts: 72
    • View Profile
Would be also great to increase the resolution to get higher render times. All high-end xeons get about the same result with the current test.

2016-03-09, 09:15:43
Reply #36

romullus

  • Global Moderator
  • Active Users
  • ****
  • Posts: 8779
  • Let's move this topic, shall we?
    • View Profile
    • My Models
I have feature request for benchmark score table - make it able to sort records by other columns, not only render time. Ability to sort by CPU would be most welcome as it would allow to estimate average performance of particular CPU.

I saw i7 4771 entry with 7:16 time, while mine does it in about 5 minutes. Without sorting list by CPU, it's impossible to hunt down all entries of certain PC configuration and thus it makes this benchmark less effective than it could be.
I'm not Corona Team member. Everything i say, is my personal opinion only.
My Models | My Videos | My Pictures

2016-03-09, 10:09:51
Reply #37

cebli

  • Active Users
  • **
  • Posts: 13
    • View Profile
    • Architectural Visualisation & Animation
is it posible to see all row in one page like benchwell? they have 1490 row and thats easy to read and find any cpu.

2016-03-09, 10:12:33
Reply #38

FrostKiwi

  • Active Users
  • **
  • Posts: 686
    • View Profile
    • YouTube
we will tomorrow probably release updated version of the benchmark that reports consistent times in the popup window (the one sent to web seems OK), and runs even if MSVS fails to install
heh - thanks.
I have 3 systems, all three for some awesome reason post VS++ 2015 install errors. Win7 64bit, 2x 8.1 64bit
all updated windows.
I found the culprit on one system, but not the other 2.

All redists of c++ 2015 fail to install, culprit on one being a broken security update, that you have to force install through Deployment Image Servicing and Management. I can't get that update installed on the other two, every fix for this is PC specific and can have a multitude of reasons, ranging from too small reserved system partition (deleted it on one PC for programming reasons on my part) to random shit like Cisco VPN clients conflicting.
There are pages upon pages of this update's clusterfuck and it forbidding to install vs c++ 2015, whilst posting awesome and very descriptive "unspecified error" messages.
Basicly an exeption in the code, with no explanation or debug messages to help.

One system is freshly installed, 100% empty - no idea why there.
I'm 🐥 not 🥝, pls don't eat me ( ;  ;   )

2016-03-09, 11:38:14
Reply #39

Ondra

  • Administrator
  • Active Users
  • *****
  • Posts: 9048
  • Turning coffee to features since 2009
    • View Profile
Rendering is magic.How to get minidumps for crashed/frozen 3ds Max | Sorry for short replies, brief responses = more time to develop Corona ;)

2016-03-09, 12:20:02
Reply #40

pokoy

  • Active Users
  • **
  • Posts: 1850
    • View Profile
The 4-cpu system ranked #22 falls behind a 2-cpu system with the same cpu... wonder what went wrong here.

Looking forward to the updated benchmark, would really like to know how my machine compares.

2016-03-09, 12:44:09
Reply #41

maru

  • Corona Team
  • Active Users
  • ****
  • Posts: 12710
  • Marcin
    • View Profile
Feature request: rays/s recorded in the table - it is proportional to time in this particular scene, but could serve as reference for other scenes (to see if there is some mistake in the scene or if cpu is performing as it should)

Bug report: splash screen is heavily JPG'd
Marcin Miodek | chaos-corona.com
3D Support Team Lead - Corona | contact us

2016-03-09, 13:01:35
Reply #42

Jann

  • Active Users
  • **
  • Posts: 142
    • View Profile
The 4-cpu system ranked #22 falls behind a 2-cpu system with the same cpu... wonder what went wrong here.

Looking forward to the updated benchmark, would really like to know how my machine compares.
Those are E5-2xxx cpus, they don't work in 4-cpu configs. So probably some kind of bug. (ES cpus, too many cores, etc.)

2016-03-09, 15:05:29
Reply #43

becki

  • Users
  • *
  • Posts: 4
    • View Profile
please send optimal settings for 3930k P9X79 32gb2133mhz

2016-03-09, 15:48:04
Reply #44

Gtelis

  • Active Users
  • **
  • Posts: 14
    • View Profile
Corona 1.3 Benchmark Finished
BTR Scene 16 passes
Intel(R) Xeon(R) CPU E5-2650 v3 @ 2.30GHz (x2)
Time: 0:01:17, Rays/sec: 6,362,200

Still waiting for the xeon E7-E5 46xx guys to put us to shame :P