Profile Picture

CC3plus Base character Render Test in Iclone 7

Posted By sungod 3 Years Ago
You don't have permission to rate!
1
2
3

CC3plus Base character Render Test in Iclone 7

Author
Message
4u2ges
4u2ges
Posted 3 Years Ago
View Quick Profile
Distinguished Member

Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)

Group: Forum Members
Last Active: 5 days ago
Posts: 5.3K, Visits: 16.7K
Very good Sungod. Glad it helped! :)
Looking forward.




sungod
sungod
Posted 3 Years Ago
View Quick Profile
Distinguished Member

Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)

Group: Forum Members
Last Active: 3 Months Ago
Posts: 249, Visits: 3.7K
Super Sampling 3x3 has noticable quality increase together with full screen optimisation for 1216 x 2160 on 4K screen.
And also for normal 4k and 1080p
Infact Supersampling  is a must for the benefits in quality vs speed.
And final renders it's now important.
1Frame per second is speed with
Super sampling MP4 Hardware encoder and 1216 x 2160 .

With this Said I will create another thread with more parameter checked as suggested by @4u2ges
sungod
sungod
Posted 3 Years Ago
View Quick Profile
Distinguished Member

Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)

Group: Forum Members
Last Active: 3 Months Ago
Posts: 249, Visits: 3.7K
@4u2ges
 I am blown by two Things Hardware encoder and Fullscreen optimisation
So quick render of 1080p on 4K screen was around 28 seconds 
Without above two settings it was 42 seconds

But as you explained 1080p rendered on 4k screen is WOW
Similarly 4k render on 4k screen is not good as compared with 1080p on 4k with fullscreen.

@4u2ges I will create another thread with updated project now to get better Benchmarking tests done now that you have rectified my mistakes and getting best quality renders.
Cheers take care
sungod
sungod
Posted 3 Years Ago
View Quick Profile
Distinguished Member

Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)

Group: Forum Members
Last Active: 3 Months Ago
Posts: 249, Visits: 3.7K
Glad to understand the quirks more thanks to 4u2ges. Yours is a 10 star post.

I didn't try optimise full screen and I will do that later today.
All my previous results were on a 4k screen. I will try optimise full screen and update and also try super sampling. 😃
4u2ges
This post has been flagged as an answer
4u2ges
Posted 3 Years Ago
View Quick Profile
Distinguished Member

Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)

Group: Forum Members
Last Active: 5 days ago
Posts: 5.3K, Visits: 16.7K
I got another one for you.
"New" users might not know and "Old" users sometimes forget...
But the screen resolution and view-port size would impact your render speed enormously and would influence the quality of the render in a way.

I got an old i7 2600K CPU, 32 GB of RAM and a pair 1080Ti (second is for Iray render only). All drives are SSD
And I ALWAYS render with Full Screen.

So those are my numbers for your project:
2K, 4K is render Frame Size
ON 2K, ON 4K, is my screen resolution
All renders with "Optimize with Full Screen Render" checked

2K ON 2K MP4 - 40 sec
2K ON 2K AVI - 43 sec
4K ON 2K MP4 - 129 sec
4K ON 2K AVI - 145 sec

2K ON 4K MP4 - 123 sec
2K ON 4K AVI - 127 sec
4K ON 4K MP4 - 92 sec
4K ON 4K AVI - 119 sec

Funny isn't that with full 4K screen 4K frame renders faster than 2K. Don's know if anyone gets the same but that's how it is on my system.

However the quality of 2K rendered on 4K is better than 2K on 2K.
Although on this particular project it is less noticeable than on some dynamic project with a lot of straight lines (like walls, buildings...) moving across the screen under certain angles.
Worse yet if you have some small GI emissive lights in various forms in a project and then they flicker like crazy on your render and you wonder why...
That is when you will learn to appreciate anti-aliasing (by enabling super-sampling, or rendering on hi-res screen and enabling "Optimize with Full Screen Render").

So the bottom line is, do not always chase for the fastest render time for every other project.
When I choose how to render and knowing about most of the quirks of RL render engine, I'd rather go for the slower, but quality render.







sungod
sungod
Posted 3 Years Ago
View Quick Profile
Distinguished Member

Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)

Group: Forum Members
Last Active: 3 Months Ago
Posts: 249, Visits: 3.7K
Same test in MP4
AMD  5950, 2080 Super Card, Win 11, Windows/IClone SSD different than Target SSD of video both WD SN850
4K MP4 576 Frames @ 96 Seconds  so 6 Frames Per second
4K AVI Uncompressed 576 Frames @ 72 Seconds  so 8 Frames Per Second

1080 Render test 
1080p Mp4 test 576 Frames @ 42 Seconds
1080p AVI uncompressed @ 42 Seconds
around 14 Frames per second.

Drive speed is a factor in 4K and non Factor in 1080p in above test
But Drive speed is absolutely a BIG factor irrespectively which is clear from test by @animagic. :)

Edit:  Looking forward to get crushed in a Heavy Scene. Will Make another thread for it later.

Enlighten me @all
sungod
sungod
Posted 3 Years Ago
View Quick Profile
Distinguished Member

Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)Distinguished Member (1.4K reputation)

Group: Forum Members
Last Active: 3 Months Ago
Posts: 249, Visits: 3.7K
animagic (2/13/2022)
I ran the test.


* Rendering to 4k PNG sequence:
Target SSD: 5:00 = 300 sec (1.92 fps)

Not so good, as you can see...:unsure: 

Anyway, this has been quite enlightening for me.


Image test for me is slow too.
@4u2ges 
You are absolutely right about SSD io speed. Maybe Animagic biggest improvement could be his SSD

Guys can we test same thing but MP4 too?
@Animagic
@yekoo
@4u2ges

4u2ges
This post has been flagged as an answer
4u2ges
Posted 3 Years Ago
View Quick Profile
Distinguished Member

Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)Distinguished Member (22.3K reputation)

Group: Forum Members
Last Active: 5 days ago
Posts: 5.3K, Visits: 16.7K
576 frames of 4K uncompressed AVI would result in Gbs of data writing.
So this is not a "true" iClone render test.
Portion of the test is... just testing your system IO speed.
Copy resulted render file to the same drive and time it.
Then deduct that time divided by 2 (given read/write is at the same speed) from iClone render test.
That would give you a more or less true GPU render speed.

You can say I will render to MP4, it would be very small render output file and does not impact IO at all.
But here you have another burden - a compression which puts a toll on a render speed (one area which might need to be improved- only RL can tell).
And again amount of VRAM does not affect render speed. It is a speed of VRAM and GPU architecture plays a major role.






yepkoo
yepkoo
Posted 3 Years Ago
View Quick Profile
Distinguished Member

Distinguished Member (3.7K reputation)Distinguished Member (3.7K reputation)Distinguished Member (3.7K reputation)Distinguished Member (3.7K reputation)Distinguished Member (3.7K reputation)Distinguished Member (3.7K reputation)Distinguished Member (3.7K reputation)Distinguished Member (3.7K reputation)Distinguished Member (3.7K reputation)

Group: Forum Members
Last Active: Last Month
Posts: 1.2K, Visits: 1.9K
Good system is very good for your work scene.
However, I opened a thread about this a long time ago.
iClone is limited by the maximum system resources it will use while rendering.
Maybe the iC8 is different but the system doesn't have to be perfect when rendering in iC7.


----------------------------------------------------------------
https://www.youtube.com/yepkoo
https://dev.epicgames.com/community/profile/bxvo/Yepkoo#learning


animagic
animagic
Posted 3 Years Ago
View Quick Profile
Distinguished Member

Distinguished Member (33.4K reputation)Distinguished Member (33.4K reputation)Distinguished Member (33.4K reputation)Distinguished Member (33.4K reputation)Distinguished Member (33.4K reputation)Distinguished Member (33.4K reputation)Distinguished Member (33.4K reputation)Distinguished Member (33.4K reputation)Distinguished Member (33.4K reputation)

Group: Forum Members
Last Active: Last Month
Posts: 15.8K, Visits: 31.3K
I ran the test.

System specs: 
Processor:  11th Gen Intel(R) Core(TM) i9-11900F @ 2.50GHz   2.50 GHz
Installed RAM:   128 GB
System type:   64-bit operating system, x64-based processor
OS:   Windows 10 Pro, Version   20H2
GPU: NVIDIA RTX 3090 
Driver version: 472.47
VRAM: 24 GB
****
Drives used in test:
SSD: Samsung 860 (or 870) EVO 4 TB drive
HD: Toshiba 2 TB drive

Project: DH Sample.iProj (as provided)

* Rendering to 4K AVI (uncompressed) @ 24 fps (576 frames)
Target SSD: 1:28 = 88 sec (6.54 fps)
Target HD:   2:50 = 170 sec (3.39 fps)
Target Network (HD): 7:09 = 429 sec (1.34 fps)

I included the network target because that is what I actually do: I render to another PC on the network (which is inefficient, it seems)

I normally don't render to 4k, but in this test with a simple scene what you render to can make quite a difference.

I primarily use image sequences in my pipeline, so I did a test with that as well:

* Rendering to 4k PNG sequence:
Target SSD: 5:00 = 300 sec (1.92 fps)

Not so good, as you can see...:unsure: 

Anyway, this has been quite enlightening for me.





1
2
3



Reading This Topic

0 active, 0 guests, 0 members, 0 anonymous.
No members currently viewing this topic!