nVidia GTX 1030 for 1080p and madVR - Page 2

tony359 · 6th September 2017, 12:01

Good point. Struggling to find a free benchmark suit though. I'll investigate.
Re. the Ryzen, I've converted an ISO file into H265 on both machines with Handbrake. The i7 using 8 threads, the Ryzen only 4. It took exactly the same time (58 minutes and some seconds).

takenori · 7th September 2017, 15:01

the free and simplest app and with online comparison would be the free ver. of 3dmark and cpu-z

tony359 · 7th September 2017, 16:32

that's what I used. The free version does not allow to upload and compare but I guess I could just compare the numbers manually, you're right.

tony359 · 7th September 2017, 20:39

Ok, my 1050Ti checks out - a it 's 5th out of 14 tests with the same system. Puzzles me that my 460 performs better on Physics then. It's SEVEN years old!

nevcairiel · 7th September 2017, 21:36

Physics is a CPU test.

tony359 · 7th September 2017, 23:47

ah, that's the trick. I assumed it referred to the nVidia Physx - I was wrong.
Ok, it makes sense now that an old i7 outperformed the cheapest Ryzen available.

Cheers.

JarrettH · 13th September 2017, 05:14

I can run NGU-Sharp medium (luma) + super-xbr (chroma), smooth motion on, debanding low (any higher eats away detail), most trade settings off - this on a GT 730 + Intel Sandy Bridge processor, going from 720p to 1080p (1.5x). I don't think you have anything to worry about. Make sure it comes with GDDR5 memory and not lots of slow memory, which is sometimes stocked, or used to be, on low-end cards.

For BD 1080p, the only thing I would do is change the chroma scaling algorithm to something high-end like NGU-AA medium or high. It won't be that taxing still. Debanding is like a <1 ms hit, smooth motion 2-3 ms. These aren't performance hungry features.

tony359 · 13th September 2017, 10:14

Thanks for your input!
I have already purchased a 1050Ti to be on the safe side and I feel happy with that. The card is running at 5% with regular movies: that gives me peace of mind and no need to worry about temperatures.
I still need to try a better Chroma scaling and I never heard of Debanding - I'll check that out.
Thx!

7th September 2017, 21:36	#25 \| Link
nevcairiel Registered Developer Join Date: Mar 2010 Location: Hamburg/Germany Posts: 10,347	Physics is a CPU test. __________________ LAV Filters - open source ffmpeg based media splitter and decoders

13th September 2017, 05:14	#27 \| Link
JarrettH Registered User Join Date: Aug 2004 Location: Canada Posts: 860	I can run NGU-Sharp medium (luma) + super-xbr (chroma), smooth motion on, debanding low (any higher eats away detail), most trade settings off - this on a GT 730 + Intel Sandy Bridge processor, going from 720p to 1080p (1.5x). I don't think you have anything to worry about. Make sure it comes with GDDR5 memory and not lots of slow memory, which is sometimes stocked, or used to be, on low-end cards. For BD 1080p, the only thing I would do is change the chroma scaling algorithm to something high-end like NGU-AA medium or high. It won't be that taxing still. Debanding is like a <1 ms hit, smooth motion 2-3 ms. These aren't performance hungry features. Last edited by JarrettH; 13th September 2017 at 05:25.

6th September 2017, 12:01	#21 \| Link
tony359 Registered User Join Date: Feb 2013 Posts: 84	Good point. Struggling to find a free benchmark suit though. I'll investigate. Re. the Ryzen, I've converted an ISO file into H265 on both machines with Handbrake. The i7 using 8 threads, the Ryzen only 4. It took exactly the same time (58 minutes and some seconds).

7th September 2017, 15:01	#22 \| Link
takenori Registered User Join Date: Nov 2010 Posts: 48	the free and simplest app and with online comparison would be the free ver. of 3dmark and cpu-z

7th September 2017, 16:32	#23 \| Link
tony359 Registered User Join Date: Feb 2013 Posts: 84	that's what I used. The free version does not allow to upload and compare but I guess I could just compare the numbers manually, you're right.

7th September 2017, 20:39	#24 \| Link
tony359 Registered User Join Date: Feb 2013 Posts: 84	Ok, my 1050Ti checks out - a it 's 5th out of 14 tests with the same system. Puzzles me that my 460 performs better on Physics then. It's SEVEN years old!

7th September 2017, 23:47	#26 \| Link
tony359 Registered User Join Date: Feb 2013 Posts: 84	ah, that's the trick. I assumed it referred to the nVidia Physx - I was wrong. Ok, it makes sense now that an old i7 outperformed the cheapest Ryzen available. Cheers.

13th September 2017, 10:14	#28 \| Link
tony359 Registered User Join Date: Feb 2013 Posts: 84	Thanks for your input! I have already purchased a 1050Ti to be on the safe side and I feel happy with that. The card is running at 5% with regular movies: that gives me peace of mind and no need to worry about temperatures. I still need to try a better Chroma scaling and I never heard of Debanding - I'll check that out. Thx!