Software Engineering Courses (SECourses)•2y ago

@Dr. Furkan Gözükara can you possibly add more information in the console during process? On runpod

@Dr. Furkan Gözükara can you possibly add more information in the console during process? On runpod i am only seeing this

rafrafOP•3/2/24, 3:19 PM

is there a way to run a tail to see more of the process?

Rrafraf @Dr. Furkan Gözükara can you possibly add more information in the console during...

Furkan Gözükara SECourses•3/2/24, 3:20 PM

requires to modify their pipeline which i dont want to

FFurkan Gözükara SECourses requires to modify their pipeline which i dont want to

rafrafOP•3/2/24, 3:21 PM

Furkan Gözükara SECourses•3/2/24, 3:21 PM

because it would make it harder to merge once they improve

rafrafOP•3/2/24, 3:21 PM

yea of course

rafrafOP•3/2/24, 3:24 PM

by the way, your gui says v16 not v17

rafrafOP•3/2/24, 3:24 PM

Furkan Gözükara SECourses•3/2/24, 3:24 PM

let me fix

Furkan Gözükara SECourses•3/2/24, 3:25 PM

please do update

rafrafOP•3/2/24, 3:25 PM

also

Dazzastrous [4090]•3/2/24, 3:25 PM

is that v18 now

Furkan Gözükara SECourses•3/2/24, 3:25 PM

v17

rafrafOP•3/2/24, 3:26 PM

i am getting OOO with an A6000 48gb. Seems like i still need the memory optimisation ??

rafrafOP•3/2/24, 3:26 PM

rafrafOP•3/2/24, 3:26 PM

rafrafOP•3/2/24, 3:26 PM

Furkan Gözükara SECourses•3/2/24, 3:26 PM

yes i know

Furkan Gözükara SECourses•3/2/24, 3:26 PM

for very high

Furkan Gözükara SECourses•3/2/24, 3:26 PM

you need 80 gb a100

rafrafOP•3/2/24, 3:27 PM

i didnt realise that

rafrafOP•3/2/24, 3:27 PM

so the A6000 is not good, because same price as 4090 but slower chip

Furkan Gözükara SECourses•3/2/24, 3:27 PM

Rrafraf so the A6000 is not good, because same price as 4090 but slower chip

Furkan Gözükara SECourses•3/2/24, 3:27 PM

it is useful for vram

rafrafOP•3/2/24, 3:28 PM

yea but but almost x2

rafrafOP•3/2/24, 3:28 PM

will it cut in half the compute time compared to 4090?

Furkan Gözükara SECourses•3/2/24, 3:28 PM

nope

Furkan Gözükara SECourses•3/2/24, 3:28 PM

for single gpu

Furkan Gözükara SECourses•3/2/24, 3:29 PM

4090 fastest

Furkan Gözükara SECourses•3/2/24, 3:29 PM

unless you get like h100

Furkan Gözükara SECourses•3/2/24, 3:29 PM

these are useful for high vram demand

rafrafOP•3/2/24, 3:29 PM

yes but A100 no optimisation vs 4090 with optimisation, which one is faster, less time, by how much? Significant?

Furkan Gözükara SECourses•3/2/24, 3:29 PM

i didnt test

Furkan Gözükara SECourses•3/2/24, 3:29 PM

give it a try

rafrafOP•3/2/24, 3:29 PM

Ok, i cant find an h100 on runpod, only on community

rafrafOP•3/2/24, 3:30 PM

and the bandwith is very low

rafrafOP•3/2/24, 3:31 PM

yea there are no h100 or a100's avaialbe at all

Furkan Gözükara SECourses•3/2/24, 3:31 PM

ye they are in super demand

Dazzastrous [4090]•3/2/24, 3:32 PM

Supir Demand

Furkan Gözükara SECourses•3/2/24, 3:32 PM

stage 1

KGTachi•3/2/24, 3:34 PM

@Dr. Furkan Gözükara seems to be broken, when ever i click apply stage 1 before stage 2, the final image is the blurry one

rafrafOP•3/2/24, 3:34 PM

found an a100

rafrafOP•3/2/24, 3:34 PM

gonna test it

KKGTachi @Dr. Furkan Gözükara seems to be broken, when ever i click apply stage 1 before ...

Furkan Gözükara SECourses•3/2/24, 3:36 PM

ye stage 1 useless

Dazzastrous [4090]•3/2/24, 3:36 PM

wait so we only use stage 2 now?

DDazzastrous [4090]wait so we only use stage 2 now?

Furkan Gözükara SECourses•3/2/24, 3:36 PM

we were always

Dazzastrous [4090]•3/2/24, 3:38 PM

so stage 2 becomes stage 1

KGTachi•3/2/24, 3:42 PM

also maybe a button to stop the batch upscaling in newer version?

DDazzastrous [4090]so stage 2 becomes stage 1

Furkan Gözükara SECourses•3/2/24, 3:44 PM

KKGTachi also maybe a button to stop the batch upscaling in newer version?

Furkan Gözükara SECourses•3/2/24, 3:44 PM

let me see if possible

@Dr. Furkan Gözükara can you possibly add more information in the console during process? On runpod

Similar Threads