NVENC on different hardware comparison

[cartman0208]

If you guys have the time, can you do encodes with the same settings as me? Constant quality 27, audio passthrough and constant framerate for video?

Sure ... 43 seconds, same as before, the resulting file is 15MB bigger though.

 

[Flymo]

Those are the setting I did....

Oops! Have you tried the slowest preset which is supposed to give better quality at the same bitrate?

 

[Flymo]

Sure ... 43 seconds, same as before, the resulting file is 15MB bigger though.

Early results show that hardware makes a virtually inconceivable difference to encoding. I think 43s vs 49s is nothing to worry about. My GPU could be throttling inside its laptop chassis.

Btw since you game, be mindful of the supposed gaming bugs in the driver you just installed.

 

[Jimc115]

If you guys have the time, can you do encodes with the same settings as me? Constant quality 27, audio passthrough and constant framerate for video?

Got the same with HandBrake....do you have any filters turned on???
The Preset has everything off.

 

[cartman0208]

Early results show that hardware makes a virtually inconceivable difference to encoding. I think 43s vs 49s is nothing to worry about. My GPU could be throttling inside its laptop chassis.

That was my thought ... your clock is at 1110, obviously the mobile models are throttled to prevent heat issues, mine was 1500, that could explain the difference of 12% in encoding time.

On the other hand ... did you use the preset I mentioned in the first post?
There's one for 1080p and one for 4K

 

[Jimc115]

Early results show that hardware makes a virtually inconceivable difference to encoding. I think 43s vs 49s is nothing to worry about. My GPU could be throttling inside its laptop chassis.

Btw since you game, be mindful of the supposed gaming bugs in the driver you just installed.

Adjusted preset down to slowest...still came in at 35 sec.

 

[Flymo]

Adjusted preset down to slowest...still came in at 35 sec.

I adjusted mine down to slowest and it took 8m42s. The quality was quite astonishing considering the resulting file size was only 745MB.

 

[Jimc115]

I adjusted mine down to slowest and it took 8m42s. The quality was quite astonishing considering the resulting file size was only 745MB.

Agreed...that's a fantastic Codec.

 

[Flymo]

Got the same with HandBrake....do you have any filters turned on???
The Preset has everything off.

I have no filters turned on. This is also a fresh download of the portable version of VidCoder since I have never used it before. It's nice that it doesn't need an install. I wonder if it needs .NET 6.0 like HandBrake does?

 

[Jimc115]

I adjusted mine down to slowest and it took 8m42s. The quality was quite astonishing considering the resulting file size was only 745MB.

8m42...you reencode the 6gb file....running it now.

 

[Jimc115]

I have no filters turned on. This is also a fresh download of the portable version of VidCoder since I have never used it before. It's nice that it doesn't need an install. I wonder if it needs .NET 6.0 like HandBrake does?

Yes...it does.

 

[Flymo]

That was my thought ... your clock is at 1110, obviously the mobile models are throttled to prevent heat issues, mine was 1500, that could explain the difference of 12% in encoding time.

On the other hand ... did you use the preset I mentioned in the first post?
There's one for 1080p and one for 4K

I did use the 1080p preset stated in your post then mistakenly used it again for the 4k encode. I then chose the 4k profile for another 4k encode but I got the same runtime anyway.

When I encode the videos, Windows task manager shows GPU video encode usage is at 100% but the 3D part of the GPU isn't being used (as expected) and although the fans did ramp up, temperatures weren't of any great concern so I wouldn't think it would be throttling. Hard to say when it comes to laptops. I'd say the 3080 165W Laptop GPU should easily beat the 1660 Ti desktop GPU in games though if that helps the situation.

 

[Jimc115]

I did use the 1080p preset stated in your post then mistakenly used it again for the 4k encode. I then chose the 4k profile for another 4k encode but I got the same runtime anyway.

When I encode the videos, Windows task manager shows GPU video encode usage is at 100% but the 3D part of the GPU isn't being used (as expected) and although the fans did ramp up, temperatures weren't of any great concern so I wouldn't think it would be throttling. Hard to say when it comes to laptops. I'd say the 3080 165W Laptop GPU should easily beat the 1660 Ti desktop GPU in games though if that helps the situation.

Yeah...overall...you benchmark 67% faster than the 1660ti Desktop.

 

[Jimc115]

Ok...6min 57sec at lowest speed. And..you're right...it's a better encode than their 700mb one. I love the NVENC Codec.

 

[cartman0208]

When I encode the videos, Windows task manager shows GPU video encode usage is at 100% but the 3D part of the GPU isn't being used (as expected) and although the fans did ramp up, temperatures weren't of any great concern so I wouldn't think it would be throttling.

I meant, the mobile models are lower clocked by default...
But interestingly, the 3D graph of my card jumps around 60 to 95% utilization during a 4K encode, during a 1080p encode more like 90 to 95%
Video encode graph is flat at 100%
The 4K encode just finished after 7m26s at slowest

Yeah...overall...you benchmark 67% faster than the 1660ti Desktop.

I do not doubt that ... but I wouldn't want to game on a Laptop

 

[Flymo]

Ok...6min 57sec at lowest speed. And..you're right...it's a better encode than their 700mb one. I love the NVENC Codec.

Technically (I may be wrong here) NVENC is just an encoder. An incredibly fast and efficient encoder. H.265 is the codec.

 

[Flymo]

I meant, the mobile models are lower clocked by default...
But interestingly, the 3D graph of my card jumps around 60 to 95% utilization during a 4K encode, during a 1080p encode more like 90 to 95%
Video encode graph is flat at 100%
The 4K encode just finished after 7m26s at slowest


I do not doubt that ... but I wouldn't want to game on a Laptop

This was my first high-end laptop (M.2 SSD + 8-core i9 + 32GB RAM + high end GPU + high end cooling). I bought it after ragequitting my plans to purchase new desktop components during the COVID price gouging days. I have the laptop plugged in to my LG 48" OLED that sits on my desk where I also watch some movies and TV shows. You'd be surprised. The laptop experience is hardly a compromise these days.

 

[Jimc115]

I meant, the mobile models are lower clocked by default...
But interestingly, the 3D graph of my card jumps around 60 to 95% utilization during a 4K encode, during a 1080p encode more like 90 to 95%
Video encode graph is flat at 100%
The 4K encode just finished after 7m26s at slowest


I do not doubt that ... but I wouldn't want to game on a Laptop

Technically (I may be wrong here) NVENC is just an encoder. An incredibly fast and efficient encoder. H.265 is the codec.

Correct...my mistake.




3D Never jumped over 1%...
Maxed Video Encode as Cartman did
Utilization stayed at between 95-100%
Looks like my GPU is stealing 64GB's of my memory though...lol
GPU stayed at about 57C

 

[Flymo]

Correct...my mistake.

3D Never jumped over 1%...
Maxed Video Encode as Cartman did
Utilization stayed at between 95-100%
Looks like my GPU is stealing 64GB's of my memory though...lol
GPU stayed at about 57C

Good lord that is a lot of memory. Good enough disks there too?

It is interesting that video encoding is at 100% for all three of us but cartman0208's 3D usage spikes heavily. It is also interesting that all of our runtimes are different yet nothing drastic so much. I assume CPU grunt shouldn't really come in to play?

 

[Jimc115]

This was my first high-end laptop (M.2 SSD + 8-core i9 + 32GB RAM + high end GPU + high end cooling). I bought it after ragequitting my plans to purchase new desktop components during the COVID price gouging days. I have the laptop plugged in to my LG 48" OLED that sits on my desk where I also watch some movies and TV shows. You'd be surprised. The laptop experience is hardly a compromise these days.

Yeah...I was worried about the price gouging also....with no onboard video and the 1060, I bought a Ryzen 7 5700G just in case that's now just sitting on the shelf still in the box.