Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
Tags

Clarification on hardware needed for 1080p videos at 60FPS, please.

A topic by stgemma created Apr 13, 2020 Views: 2,954 Replies: 12
Viewing posts 1 to 9
(2 edits)

I was wondering what is the hardware necessary to get 60fps at 1080p for animations, like those in the example videos seen on the page where you can download the app?

This gtx 1050 2gb I have seems to not be able to cut it for anything, except to render a short gif with the results of around 200p, which is useless, because it's so tiny.

Developer

Without the "Split" option enabled, i think there is no current graphic card that can handle 1080p.

The output fps don't use more memory from your card.

What about nvideas quadro cards, the 8000 for example, are they not able to handle this?

Developer

There was a lot of memory improvements in those two months, and should be a few more on the way. So its a little hard to tell right now.

I see, thank you. 

Which graphics card did you all use to make those 1080p, 60fps example videos?

Grisk is right even with a 2080 SUPER you can't slow down 1080p videos.

Wow. What kind of graphics card would it require for such a task?

(1 edit)

This is my (somewhat costly) solution to creating 1080p or 4K 60fps videos using DAIN-App:

If your video is native 1080p, downscale the video first to 848×480. I use a bicubic sharp filter to downscale the video in avisynth. This should retain as much detail as possible.

Next run the video in Dain-App. Converting should be much faster than using the Split method. 

Here comes the somewhat costly part: Buy Gigapixel (The free beta is sadly over) for Video to reconstruct the video back to 1080p or make it 4K. 

Basically you are using 2 AI programs to create your ideal video creation. Hope that helps.

I have a RTX 2060 + Ryzen 7 CPU and converting videos @720p is way too slow for anything over 5 minutes. AI upscaling is the only way IMO.

(1 edit)

Cool beans! Thank you for that information. That is helpful.

Using RTX 2080 Ti 11 GB + Ryzen 9 CPU 3900 and 700 section size in split frame setting,  it can render 12 interpolated frames in one minute for 1080p 24 fps video into 60 fps. 

(1 edit)

Win 10 /// RTX 2080 Ti Strix OC 11Gb (water cooled) / Ryzen 3970x / 128 RAM / m.2 970 Evo Plus SSD

Test file 0,5sec 1080p 30fps x2 interpolation (15 frames into 29). 15 frames original from the video. This means that only 14 new frames was generated.
Split frame into section setting:
 @ 850x150 - 3m:03s ~ 12 frames rendered about 2m:37s
 @ 700x150 - 3m:26s ~ 12 frames rendered about 2m:56s

Even if calculate the processing of all 29 frames during interpolation, your PC configuration completed this task faster.
Maybe it depends on the content in the image? Otherwise, I don’t understand why my computer rendered frames much slower than yours ...

(1 edit)

Recently, I try to render 1920x1080 film from 24fps to 144fps (interpolate 8x setting)

1%|         | 906/90011 [15:31:38<6264:55:24, 253.11s/it, file=0000102728.p

1%|        | 907/90011 [16:06:56<20110:57:51, 812.53s/it, file=0000102811.p

......

1%|        | 912/90011 [17:04:47<24265:53:08, 980.45s/it, file=0000104146.p

1%|       | 913/90011 [18:15:10<48336:48:31, 1953.05s/it, file=0000105397.p

1%|       | 914/90011 [18:19:23<35713:04:01, 1443.00s/it, file=0000107065.p  

1%|       | 915/90011 [18:21:27<25925:17:02, 1047.53s/it, file=0000107149.p  

....

1%|         | 922/90011 [18:36:13<5016:15:13, 202.70s/it, file=0000107441.p  

1%|         | 923/90011 [18:38:22<4463:26:52, 180.37s/it, file=0000107482.p  

I find that the processing time is varying from 180s/it to 1953s/it. 

I get about the same speed with my 2080ti + 3950X