@johnebaker

Hi John.

This is the MdediaInfo from my HEVC render from VPX 14 following the instructions I gave earlier.

General
Complete name                            : J:\2 Minute Tech Tips Audio Master VPX HEVC.MP4
Format                                   : MPEG-4
Format profile                           : Base Media
Codec ID                                 : iso4 (iso4/hvc1)
File size                                : 610 MiB
Duration                                 : 2 min 21 s
Overall bit rate mode                    : Variable
Overall bit rate                         : 36.1 Mb/s
Encoded date                             : UTC 2023-02-19 11:38:23
Tagged date                              : UTC 2023-02-19 11:38:23

Video
ID                                       : 1
Format                                   : HEVC
Format/Info                              : High Efficiency Video Coding
Format profile                           : Main@L5.1@Main
Codec ID                                 : hvc1
Codec ID/Info                            : High Efficiency Video Coding
Duration                                 : 2 min 21 s
Bit rate                                 : 35.9 Mb/s
Width                                    : 3 840 pixels
Height                                   : 2 160 pixels
Display aspect ratio                     : 16:9
Frame rate mode                          : Constant
Frame rate                               : 50.000 FPS
Color space                              : YUV
Chroma subsampling                       : 4:2:0
Bit depth                                : 8 bits
Bits/(Pixel*Frame)                       : 0.087
Stream size                              : 607 MiB (99%)
Language                                 : English
Encoded date                             : UTC 2023-02-19 11:38:23
Tagged date                              : UTC 2023-02-19 11:38:23
Codec configuration box                  : hvcC

Audio
ID                                       : 2
Format                                   : AAC LC
Format/Info                              : Advanced Audio Codec Low Complexity
Codec ID                                 : mp4a-40-2
Duration                                 : 2 min 21 s
Bit rate mode                            : Variable
Bit rate                                 : 196 kb/s
Channel(s)                               : 2 channels
Channel layout                           : L R
Sampling rate                            : 48.0 kHz
Frame rate                               : 46.875 FPS (1024 SPF)
Compression mode                         : Lossy
Stream size                              : 3.31 MiB (0%)
Language                                 : English
Encoded date                             : UTC 2023-02-19 11:38:23
Tagged date                              : UTC 2023-02-19 11:38:23
mdhd_Duration                            : 141525

Is it possible this is a something else causing this? I'm not on Win 11 for instance.

Ray.

@CubeAce

Hi Ray

. . . . Is it possible this is a something else causing this? I'm not on Win 11 for instance.. . . .

Causing what?

The only differences I can see between what I did and the above are:

• a question of terminology - I interpreted 4K as being 4K 4096 x 2160 whereas you have exported as 4K UHD 3840 x 2160.
   
• As far as the bitrate differences are concerned you can ignore these - I was concentrating on whether 4K 50fps, and higher, was possible - the bitrate differences are due to the quick test I did was on my laptop with a low action video clip - I did hide the bitrate in the video section and forgot the top section 😕

John

@johnebaker

Hi John.

a question of terminology - I interpreted 4K as being 4K 4096 x 2160 whereas you have exported as 4K UHD 3840 x 2160.

Good point.

I have amend the Benchmark Topic to include that it is UHD. I forget that we all miss information glancing at images and text and it is sometimes best to have both. The settings for 4K (UHD) were shown in the images but hay-ho, I understand anyone reading 4K would make an assumption I put the correct terminology in text and not look at the images. My bad, but learning, I hope.

Ray.

@CubeAce

I added to my result that I have "Resizable BAR (rBAR) =ON".

In the following link I recorded the EXPORT with ShareX.
Then I show the task manager.

With screen capture software running in the background, the export takes about 43 seconds longer.

https://drive.google.com/file/d/1VMX3OLuqmOHbouUdgFP1rBHA29NRKoXT/view?usp=share_link

@Marc-Goder

Hi.

Interesting video.

It shows me from what I can see that as well as the GPU's influence that even a slower CPU clock speed with more cores pressed into service seems to be of benefit. Whether a CPU with less cores would press more of the cores to do more to compensate I don't know. The slow down caused be the first fifteen seconds and last eight and a quarter seconds of animated graphics seems comparatively the same for all of us.

I also noticed you are not using the latest nvidia drivers and you have a disabled Intel HD 630.

If both of those things are now of benefit, it questions the quoted specs from Magix perhaps? At least for newer components.

I understand the additional problems associated with screen recording.

Thank you for sharing.

Ray.

 I bought what was top of the list, biggest CPU, most amount of RAM & biggest GPU, there were a few bigger items I could have got, there was a 48GB GPU, ....

That's the exact reason why I started this thread asking for some benchmark.

I'm about to build a new PC, I have quite a budget to spend. But obviously, the last thing I would want, is seeing performance degrading after buying newest hardware for a lot of $$$.... That would be somewhat... errr... disappointing.

I do hope @Gid, that you'll be able to trace down the reason for the slowness of your kit. It might be as simple as some setting in the bios or OS slowing you down on a BUS or somewhere.

@MarcinKrk

I do hope @Marc-Goder, that you'll be able to trace down the reason for the slowness of your kit. It might be as simple as some setting in the bios or OS slowing you down on a BUS or somewhere.

I think you mean Gid

@MarcinKrk

Hi.

@Marc-Goder is correct. That was a comment made by @Former user.

The problem with any benchmark is it only shows you the results for the benchmark taken.

In this instance I had decided to put forward a project than only uses the very basic of what can be found in both VPX and Movie Studio (formerly known as Movie Edit Pro), to see see how any system would cope with just the basics of the program itself. I haven't used all the effects available as that way lies madness, not to mention very long export times. I deliberately have left out any effects that have not been available to earlier versions of the program as far as I was able to.

Third party plugin effects have also been left out as not everyone is likely to have access to all of the ones I have.

I have made deliberate restrictions so any advantage gained from one system over another should be annulled. Or at least try to make the comparisons as equal as I could.

Therein lies the a dilemma. That restricts the possible workflow any one individual could use or was needed from project to project from the same user.

I know for a fact that the systems from Gid, John Brown, or John Baker would trash my system both for playback smoothness and export rendering once some of those third party plug-ins were used. Some plug-ins use more CPU cores while others use the power of the graphics cards. There is little way of telling until one is used which way that will go.

So in my opinion (and quite probably only my opinion) it is better to start off with an idea of what your personal editing needs are regarding workflow, resolutions you hope to work at and how many effects, tracks, etc you are likely to want to put into a project. What about future proofing? Not only will the system requirements go up over each new release of the program or operating system but also for a newer camera or video monitor with quicker refresh rates if recording gaming footage may in the future require higher standards of hardware.

Some people just vlog or make holiday videos. Others like to add visual effects or additional graphics or green screening. Some have very demanding camera or game footage.

No-one here or elsewhere knows what your eventual demands may be or how long you expect the new build to cope or last for. John Baker tends to upgrade his hardware more often than most others here. I buy bits and pieces as needed as I'm OK with building my PC. Do you want a laptop or a tower? Each little gain in performance can leap the cost of a computer a lot.

I think it is better to have an idea of what you want to achieve, armed with information of what you are already doing, why you feel you need to upgrade, and ask on a forum such as this one for suggestions on a computer build or suitable prebuilt machine. Then leave it to the members to argue it out and maybe come to a consensus built from their own experiences of trying to interact and use the programs. I have myself taken much advice in the past from John Baker which has proven to be correct up to now. But as we see now, there are still mysteries we may never get to the bottom of when trying to find out how a given program interacts with a given set of hardware or project settings.

Ray.

@johnebaker

Did you also connect your laptop to the power supply during the test?  Without power, my laptop's performance drops by 40%.

I would argue here a bit:

Comparing 2 sets, where one of them, on paper, is newer and few times faster / better then the other, you would expect to see at least some improvement in the performance, regardless to the scenario. I thikn it shouldn't really matter how the specific project is setup how complex transitions are, whether or not you use the greenscreen and how many movies on the timeline shows up. I think in each of those cases, more powerful desktop should simply do better, then the other one. Not necessarily significantly better, but at least should be no worse!

So either - there's actually some "incompatibility"? of a specific GPU with the MAGIX software, or there's some overlooked setting in BIOS / System / MAGIX which is giving this powerful gear a hiccup. Afterall Gid was getting significantly poorer results then others (despite the dream-setup he has).

So that's why I really keep my fingers crossed for finding the solution with @Gid hardware, as if I I'm going to spend few thousand $$$ on the newest gear, I'd rather not get disappointed. And I'd love to refresh my gear in the coming weeks :)

@Marc-Goder

Hi

. . . . Did you also connect your laptop to the power supply during the test? . . . .

I did, in my post I said 'plugged in' = Reingesteckt oder An Strom angeschlossen

John EB

@MarcinKrk

Do you play a lot of games too?

I don't play computer games.

I use the computer to use the Internet, photo editing, typical office applications and video editing.

Your computer seems to be so good that you can wait a little longer.  If you wait until the next Magix version to buy the hardware and see how the integration of the Intel ARC graphics cards goes, that could be an advantage.

I've seen tests where the Intel ARC graphics cards outperform all other NVidia cards and AMD Radeon cards when it comes to video editing. But at the moment any software still has bugs when it comes to the Intel Deep Link Hyper Encoder.

And by that I also mean the official programs released by Intel, such as DaVinci Resolve, Magix Video ProX, Handbrake, FFMPEG, etc. But when Hyper Encoder worked, it was unbeatable.  With any material that uses HEVC, AVCHD and AV1.

@johnebaker

Thanks, Google Translate translated it differently for me.

Google then translate " There is no high performance mode connected."

@CubeAce and others,

Hi,N

I found that what I used in the Device Options for Processing and Export made a significant difference. NVIDIA for both gave me the best results as the Intel seemed to be used more heavily, sharing the load. I suggest that others check this. The other options are Intel/NVIDIA, NVIDIA/Intel, and Intel/Intel. With Intel, NVIDIA was barely used.

John CB

Already a little few users who participate and too few results to analyze more precisely.

Unfortunately nobody has the hardware of the latest generation. (Intel 12XXX and 13XXX)

Late to the party...as always.... but I want in!! I read the requirements and conditions, although I only have MEP2022

@Reyfox

I also only have MEP2022 (2.138), didn't prove to be a disadvantage.

I await the testing files then.

@Reyfox

Hi. Link Sent.

I also used MEP 2022. Any version used is welcome.

Ray.

@CubeAce, @browj2, @Marc-Goder, @Reyfox, @MarcinKrk

Hi

A comment was made in the other topic for the results of testing about the NVidia drivers versions were are using to the effect that the 497.series were faster than the 5xx series of drivers - due to removal of the datamining (?) in the 5xx series - in my case the 528.49 driver.

After testing this I got the following results with different projects I have, from simple video + images, transitions and minimal effects to complex projects with LUT colour correction, collages, PIP etc.

Taking the 497.29 drivers as the reference point with multi GPU (my default) selected in the advanced export (4K UHD h.264/AAC)

The 528.49 driver multi GPU took 2.8% longer on average exporting.

John EB

The 528.49 driver multi GPU took 2.8% longer on average exporting

I would have guessed even more.
I didn't measure it though.

Hmm... curious. I am running an AMD driver from almost a year ago because of compatibility with some plugins. I should test and see if the latest drivers will do "anything".