• AnyStream is having some DRM issues currently, Netflix is not available in HD for the time being.
    Situations like this will always happen with AnyStream: streaming providers are continuously improving their countermeasures while we try to catch up, it's an ongoing cat-and-mouse game. Please be patient and don't flood our support or forum with requests, we are working on it 24/7 to get it resolved. Thank you.

What can be expected from hardware acceleration

The upcoming version of CloneBD will support various GPU acceleration modes.

I've run some benchmarks with an internal test version, that Elby provided us with:

- Test bed: Intel i7-6700 with integrated Intel HD 530 GPU
- Source and destination both SSD to reduce I/O impact on speed
- AVC 1080p to MKV 1080p (compressed down to ~75%), 1:59h video duration

  • "conventional" transcoding in software mode:
    92 fps average @ 90% CPU, total duration: 31 min

  • acceleration with on-board Intel HD530:
    209 fps @ 24% CPU, total duration: 13 min

  • acceleration with nVidia GeForce GT 740:
    123 fps @ 7% CPU, total duration: 23 min

  • acceleration with nVidia GeForce GTX 950:
    308 fps @ 17% CPU, total duration: 9 minutes!

I'm looking forward to testing between my Intel Broadwell i7-6950x and Zotac nVidia Geforce GTX 1080 Amp Edition. I just hope the HEVC crash I'm having gets fixed also. Here's some conversion data for the Broadwell with 20 threads for comparison:

- Test bed: Intel 17-6950x with no integrated graphics, Zotac Geforce GTX 1080 Amp Edition
- Source and destination: Samsung 950 Pro NVMe m.2 512 GB OS/Boot SSD
- AVC 1080p/TrueHD to MKV 1080p .264/AAC vbr target 3.40 mbps, 31.81 GB to 2.77 GB. 1:37:23 duration, 140093 frames total

  • conventional transcoding
    316.24 frames/sec @81% CPU, total duration: 7:23, peak frames conversion rate in content to 336 fps and in titles to 377 fps
How long till this update arrives now that you have our attention? :)

Edit: I may be running in to a bottleneck with the SSD not keeping up, a limitation in the software or some other issue as I can't seem to get a better rate no matter from what SSD to what SSD I use including the considerably slower Mushkin Eco SATA 3 to Intel SATA 3 non-OS/boot SSD's that completed in 7:24 also at 81% CPU utilization. Wish I had another m.2 non-OS/Boot SSD to test.
 
Last edited:
@Reto
I haven't followed the developement in recent years.
How far (& willing) are Intel/Nvidia/AMD with GPU accelerated OpenCL ?
 
Very interested in this subject. I know the post states start date and time but are there any other posts to read leading up to Thursday? How was the subject of an upcoming version of CloneBD learned?
 
This is what I've been waiting for. I actually e-mailed Elby a while back this year to put GPU hardware acceleration in CloneBD. They said they'd pass it to the developers my e-mail. I have an AMD FX- 8350 Eight Core CPU at 4GHz per core. And I have 32GB Corsair Vengeance DDR3 1,866MHz RAM, Windows 10 Pro 64-bit, two Pioneer BD-R 2209 16x write speed bluray burners that burn the brand Verbatim BD-R 25GB at 16x with no playback problems. My video card is an MSI GeForce GTX 980TI 6GB video card. Vey happy and looking forward to see how fast CloneBD can use my video cards power with a ripped bluray by AnyDVD HD. I only use the best so I use AnyDVD HD, CloneDVD2, CloneBD and Nero Burning ROM 2016.
 
Hardware acceleration has been on the to do list since day one, there just isn't any point in adding it unless the software is relatively stable/complete. You only use the best? Then what's Nero 2016 doing on that list. Ever since Nero 9 it all went down hill into 1 major piece of bloatware. Last time I used that was 8025 or something. After that I replaced that entire thing with ImgBurn, never looked back.
 
I have to agree - Nero really needs to be crossed off that list, it spoils the average quality score dramatically ;-)

The GTX 980 is probably way beyond the threshold of where any further speed-up is to be expected. Even when converting from SSD to SSD the drive will be the limiting factor while the GPU will be idling around (I'm exaggerating a bit, but that's basically how it is).
I hope, they're going to come out with a beta next week, they were planning to anyway. A lot of testing required.
 
i got plenty of new discs to test with, along with a 1080 ^^
 
As do I. Minus the 1080 as I don't have the cash to get one.
 
Well the OC one i ordered would still be on pre-order at least untile Q1 next year, the Advanced model they could get in a day or 2. I was done waiting, and swapped to the A model. Only slightly underclocked and they refunded the difference. At least now i can game properly. Finished Doom 2016 yesterday on 1080p ultra awesomeness ^^
 
Pete: Question about what (movie title?) was transcoded in 9 minutes (please). More importantly, what was the size of the file before transcoding and what was the size of the file after transcoding? I have had 2 hour movies which didn't need transcoding to fit on a single layer disc. My curiosity has been killing me.
Thank you for the info!
 
Pete: Question about what (movie title?) was transcoded in 9 minutes (please). More importantly, what was the size of the file before transcoding and what was the size of the file after transcoding? I have had 2 hour movies which didn't need transcoding to fit on a single layer disc. My curiosity has been killing me.
Thank you for the info!

If course, I was talking about actual compression (BD to MKV).
 
Sorry Pete, I'm an old fart and I had no intention of implying you were not talking about compression. Its just that I have gotten into a mode of thinking file size before and after transcoding to give me a better sense what was accomplished by the hardware acceleration.
 
Sorry Pete, I'm an old fart and I had no intention of implying you were not talking about compression. Its just that I have gotten into a mode of thinking file size before and after transcoding to give me a better sense what was accomplished by the hardware acceleration.
I see what you mean - I didn't see what relevance the file size would have.
I assume, it's the intuitive thought, that compression would require more computing power, when the output is to be smaller (or larger? I believe both would make some sense).
But in reality it doesn't really make much of a difference at all.

The only thing that matters is: was the file just copied or transcoded (as in: decoded and then re-encoded, which is the part that makes the chips heat up - even if you don't change the file size).
 
Back
Top