What dictates Video Preview Quality?

Kit wrote on 11/30/2014, 7:41 PM
This is a novice question but what effects video preview quality? How to have it set to best full and not have jerky playback? . What makes the biggest difference? Video card, ram, processor, preview window size, motherboard? Or is it just the number of tracks and the FX applied. I'm building a title sequence with 1o tracks fading in and fading out and I can't judge what is happening as it is so jerky. Even with draft mode auto that doesn't seem to help much. Rendering down to a single track takes too long to make it an efficient way of working. Is my PC just not powerful enough? Insights welcome, thanks.

==============================================
Windows 7 64 bit, 16 GB ram
GeForce GTX 550 Ti v334.89
Processor: Intel(R) Core(TM) i7-3770K CPU @ 3.50GHz, 3501 Mhz, 4 Core(s), 8 Logical Processor(s)
Asus P8Z77-V Pro Motherboard
BIOS: American Megatrends Inc. 1504, 01/08/2012
Power: Corsair AX850
==============================================

Comments

OldSmoke wrote on 11/30/2014, 8:14 PM
From my own experience, for HD material 1 core per track is required for good playback quality. On top of that, AMD cards have the upper hand when it comes to playback. I speak from experience as I recently changed from 2x GTX580 to 2x R9 290. I also increase my RAM to 32GB to help with 4K editing.

Edit: When I mean cores, I mean full cores, not logical cores.

Proud owner of Sony Vegas Pro 7, 8, 9, 10, 11, 12 & 13 and now Magix VP15&16.

System Spec.:
Motherboard: ASUS X299 Prime-A

Ram: G.Skill 4x8GB DDR4 2666 XMP

CPU: i7-9800x @ 4.6GHz (custom water cooling system)
GPU: 1x AMD Vega Pro Frontier Edition (water cooled)
Hard drives: System Samsung 970Pro NVME, AV-Projects 1TB (4x Intel P7600 512GB VROC), 4x 2.5" Hotswap bays, 1x 3.5" Hotswap Bay, 1x LG BluRay Burner

PSU: Corsair 1200W
Monitor: 2x Dell Ultrasharp U2713HM (2560x1440)

videoITguy wrote on 11/30/2014, 8:17 PM
There is nothing mysterious here - you answered yourself with all of the above - so you got that right after all.

Let's assume you have a medium built PC with fast enough card and harddrive handling for multistreams concurrent.
Then what is what? ALL DEPENDS
1) What is your source format codec? what container, what bitrate, what filesize, how have you made subclips out of one event piece and how many subclips, then how many discrete events per timeline track, how many concurrent tracks????
2) Now FX processing - complex multi-layered or chained?
3) How much runtime length given in a single project?

Most all of this is very solved by using appropriate digital intermediates - how you do that ALL DEPENDs - aha, you guessed it - refer to above.
John_Cline wrote on 11/30/2014, 8:23 PM
Vegas will vary frame rate to maintain a given image quality, on the other hand, Premiere varies the image quality to maintain a given frame rate. There are advantages to both approaches, perhaps it would be nice to be able to choose one or the other at will.
Kit wrote on 11/30/2014, 9:46 PM
Thanks, so with my processor I should avoid having more than 4 tracks for optimum playback quality?
OldSmoke wrote on 11/30/2014, 10:31 PM
Kit

yes but only if you are doing picture in picture or compositing. I have always several tracks but only one is playing at a time or transitioning to another. I did a project in HD 1080 60p with the screen tiled 4x4=16 picture in picture tracks. From that project I noticed that once I added more then 6 tracks things started to become slower until very slow; I run a 3930K which is a 6core CPU.

Proud owner of Sony Vegas Pro 7, 8, 9, 10, 11, 12 & 13 and now Magix VP15&16.

System Spec.:
Motherboard: ASUS X299 Prime-A

Ram: G.Skill 4x8GB DDR4 2666 XMP

CPU: i7-9800x @ 4.6GHz (custom water cooling system)
GPU: 1x AMD Vega Pro Frontier Edition (water cooled)
Hard drives: System Samsung 970Pro NVME, AV-Projects 1TB (4x Intel P7600 512GB VROC), 4x 2.5" Hotswap bays, 1x 3.5" Hotswap Bay, 1x LG BluRay Burner

PSU: Corsair 1200W
Monitor: 2x Dell Ultrasharp U2713HM (2560x1440)

Mindmatter wrote on 12/1/2014, 4:18 AM
OldSmoke, could you elaborate on exactly how the change to the R89 290 improved your payback? Given that the 580 is a few years old if I'm not mistaken, are they noticably faster by architecture or simply because they are newer, faster cards?
Also, how exactly do you make use of two simultanous cards - one for preview only and the other for the work screen?

Thanks!

AMD Ryzen 9 5900X, 12x 3.7 GHz
32 GB DDR4-3200 MHz (2x16GB), Dual-Channel
NVIDIA GeForce RTX 3070, 8GB GDDR6, HDMI, DP, studio drivers
ASUS PRIME B550M-K, AMD B550, AM4, mATX
7.1 (8-chanel) Surround-Sound, Digital Audio, onboard
Samsung 970 EVO Plus 250GB, NVMe M.2 PCIe x4 SSD
be quiet! System Power 9 700W CM, 80+ Bronze, modular
2x WD red 6TB
2x Samsung 2TB SSD

PeterWright wrote on 12/1/2014, 5:38 AM
I don't know which variable makes most difference Kit, but I'm currently editing a 6 camera multicam project with the same CPU and getting full fps at Best Quality, and the footage is a mixture of 1 x EX1 MXF at 1920 x 1080, 2 x HDV at 1440 x 1080 , these three cams at 50i, and then 3 Sony Action Cams shooting at 1920 x 1080 but 30fps.

Spec:
CPU i7 3770 @ 3.50 Ghz
GPU: GTX 670 (Drivers 9.18.13,4411)
Ram 16 Gb

Edit - Just looked at another variable. With GPU acceleration turned OFF, playback with 6 cams went from 25 to 18 fps.
OldSmoke wrote on 12/1/2014, 7:03 AM
Peter

Multicam isnt the same as compositing or picture in picture. In multicam you still only playback one stream at a time. In compositing you are playing back at least two streams at the same time.

Proud owner of Sony Vegas Pro 7, 8, 9, 10, 11, 12 & 13 and now Magix VP15&16.

System Spec.:
Motherboard: ASUS X299 Prime-A

Ram: G.Skill 4x8GB DDR4 2666 XMP

CPU: i7-9800x @ 4.6GHz (custom water cooling system)
GPU: 1x AMD Vega Pro Frontier Edition (water cooled)
Hard drives: System Samsung 970Pro NVME, AV-Projects 1TB (4x Intel P7600 512GB VROC), 4x 2.5" Hotswap bays, 1x 3.5" Hotswap Bay, 1x LG BluRay Burner

PSU: Corsair 1200W
Monitor: 2x Dell Ultrasharp U2713HM (2560x1440)

OldSmoke wrote on 12/1/2014, 7:12 AM
The reason the R9 290 is faster is the better implementation of OpenCL in the ATI/AMD cards; Nvidia cards are more focused on CUDA. OPenCL is what is used for playback, there is no CUDA involved in playback at all.
Also the R9 290 has 4GB, PCIe 3.0 and I think evena bigger memory bandwith, 512 vs 384 on the GTX580.

One thing you have to be aware when switching to ATI/AMD: MC AVC and Sony AVC do not support the later GPUs of the ATI/AMD cards, I think it end at the 7xxx series. So even if you select within the MC AVC render template to use OpenCL it doesnt work and it will run on CPU only.

Another plus point is that many 3rd party plugins are OpenCL based and you will see great improvments for things like NewBlue and Boris FX porducts.

I have a 4 monitor setup, 3 1920x1080 computer monitors and one 42" LCD HDTV for preview. I have balanced the load between the two cards by using only ports on each. If you want to use 2 crads, make sure your system does actually support 2x PCIe x16, only socket 1366 and 2011 can do that.

Proud owner of Sony Vegas Pro 7, 8, 9, 10, 11, 12 & 13 and now Magix VP15&16.

System Spec.:
Motherboard: ASUS X299 Prime-A

Ram: G.Skill 4x8GB DDR4 2666 XMP

CPU: i7-9800x @ 4.6GHz (custom water cooling system)
GPU: 1x AMD Vega Pro Frontier Edition (water cooled)
Hard drives: System Samsung 970Pro NVME, AV-Projects 1TB (4x Intel P7600 512GB VROC), 4x 2.5" Hotswap bays, 1x 3.5" Hotswap Bay, 1x LG BluRay Burner

PSU: Corsair 1200W
Monitor: 2x Dell Ultrasharp U2713HM (2560x1440)

Mindmatter wrote on 12/1/2014, 11:31 AM
Thanks for the infos OldSmoke.

AMD Ryzen 9 5900X, 12x 3.7 GHz
32 GB DDR4-3200 MHz (2x16GB), Dual-Channel
NVIDIA GeForce RTX 3070, 8GB GDDR6, HDMI, DP, studio drivers
ASUS PRIME B550M-K, AMD B550, AM4, mATX
7.1 (8-chanel) Surround-Sound, Digital Audio, onboard
Samsung 970 EVO Plus 250GB, NVMe M.2 PCIe x4 SSD
be quiet! System Power 9 700W CM, 80+ Bronze, modular
2x WD red 6TB
2x Samsung 2TB SSD

johnmeyer wrote on 12/1/2014, 12:09 PM
How to have it set to best full and not have jerky playback? . What makes the biggest difference? Your actual question is actually more about video playback speed (i.e., sustaining full frame speed during playback) than it is about quality (spatial quality, with the full sharpness of the original).

Here's a link to some advice I have given on many previous occasions (this question has been asked before):

Four rules for fast preview speed

Also, if you search, you will find several dozen threads with additional advice.

Here are links to other threads where I gave this same advice. I am providing these in order to give you links to other threads where people were having the same, or similar problem:

Vegas Pro 10: 720p=ok, 1080p=Choke

vegas preview quality - still not getting it

Please critique my AVCHD-to-DVD production. :-)

Found One Cause of Slow Playback Speed

MultiCamera HELP - 6 Tracks of Jerky Video

OT: laptop speed
TeetimeNC wrote on 12/1/2014, 1:03 PM
Kit, I've never tried this, but depending on what your title sequences consist of you might be able to get the speedup you need by using proxies for your media. Vegas 12 and newer has proxy support built in, and there are 3rd-party tools like Vegasaur that can add it too.

Others who have tried it may provide better insight into whether this is worth trying.

/jerry
videoITguy wrote on 12/1/2014, 1:17 PM
I refer to my earlier post above...and while I mention appropriate digital intermediates as the cure for 90% of the ailments brought up by the OP...I did neglect the topic of proxies. Indeed if the problem is in the source codec/container type and number of subclips split out of events plus total number of events - then the issue goes directly to proxie usage.

However, this may put you in a real workflow bind if you are doing multiple concurrent tracks in a complex composite - then the DI is the bigger answer.
johnmeyer wrote on 12/1/2014, 2:17 PM
+1 to intermediates and proxies.These are by far the best way to get great timeline performance. Proxies will give you the smoothest playback, but you will of course lose spatial quality while you are editing.
PeterWright wrote on 12/1/2014, 5:34 PM
OldSmoke - I may have misunderstood the OP's question, but surely when I'm looking at six cameras at the same time, Vegas is playing them all back simultaneously.
OldSmoke wrote on 12/1/2014, 5:56 PM
Yes you do but at a reduced resolution for each of them.

Proud owner of Sony Vegas Pro 7, 8, 9, 10, 11, 12 & 13 and now Magix VP15&16.

System Spec.:
Motherboard: ASUS X299 Prime-A

Ram: G.Skill 4x8GB DDR4 2666 XMP

CPU: i7-9800x @ 4.6GHz (custom water cooling system)
GPU: 1x AMD Vega Pro Frontier Edition (water cooled)
Hard drives: System Samsung 970Pro NVME, AV-Projects 1TB (4x Intel P7600 512GB VROC), 4x 2.5" Hotswap bays, 1x 3.5" Hotswap Bay, 1x LG BluRay Burner

PSU: Corsair 1200W
Monitor: 2x Dell Ultrasharp U2713HM (2560x1440)

Kit wrote on 12/1/2014, 6:36 PM
Thanks , Johnny, for taking the time to put together the links. My question is about how to get both full frame speed and spatial quality - that's what I meant by "best full and not have jerky playback" - but I lacked the terminology.

From what I've read following your links I tried setting the deinterlace method to interpolated. You suggested unchecking it but I need to keep Scale Video to fit preview window as otherwise I can only see a fraction of the video. I also checked adjust size and quality for optimal playback. Results are a little better but still not as smooth as I really need.

I'm wondering now about using proxies but don't know anything about what it entails. It doesn't help that help keeps crashing Vegas. Thanks again.
Kit wrote on 12/1/2014, 6:39 PM
Thanks, what do you mean by DI? Is making a proxie the same as rendering to a new track? Sorry, I'm still getting to grips with terminology.
johnmeyer wrote on 12/1/2014, 7:36 PM
Thanks, what do you mean by DI? Is making a proxie the same as rendering to a new track?"DI" is a Digital Intermediate, or just "intermediate."

Proxy and Intermediate are two very similar ways to improve the editing experience, especially the preview performance. In both cases, you create a new video file from each and every source video file that you want to edit. This, of course, takes time to do, and consumes additional disk space. You then edit with these new files instead of the original video files from your camera.

Intermediate files have the identical properties to your original video file: they are the same resolution, same frame rate, and virtually the identical quality (the quality differences should be imperceptible to most people). However, each frame is compressed without reference to any of the previous or future frames. This is quite different from MPEG-2, HDV, AVCHD, and most other camera video formats. All of these use a compression technique which requires referencing as many as a few dozen adjacent frames in order to display each and every frame. It takes time for the video editing program (i.e., Vegas) to open all these additional frames and compute what the pixels should look like for the current frame.

The downside of intermediate video formats is that because each frame is compressed all by itself, the size of the video file is much, much larger, usually by a factor of five to ten. For instance, Cineform is a good intermediate codec. 1920x1080 video encoded using Cineform consumes between 58 and 86 GB per hour, depending on the quality settings you choose. By comparison, the AVCHD that your video camera creates probably only consumes about 8-10 GB per hour.

A proxy video file sounds like the same thing as an intermediate file, but the key difference is that it is created at a much lower resolution and quality. The file size is much smaller than an intermediate, and usually smaller than even the original video file. Because it too compresses each frame without reference to adjacent frames, it is also much faster when previewing on the timeline, and because it uses much lower resolution than the original video file, it is even faster on the timeline than an intermediate.

When you edit using a proxy, that proxy must be replaced with the original video files, just prior to rendering. There are lots of scripts that handle this for you automatically if you are using older versions of Vegas, but the current Vegas version supports proxy editing directly (i.e., it will do the replacement for you, just prior to rendering).

So, which to use, intermediate or proxy?

In general, use an intermediate when disk space is not a problem, and when you want the simplest workflow. Use proxy files if you have a lesser computer that can't give you a fast preview, even when using an intermediate. Also use a proxy instead of an intermediate if you don't have much disk space. Finally, even with a fast computer, if you add a lot of effects or apply other advanced editing to your video, you may find that preview speed becomes a problem, even with an intermediate. This is another reason why you might want to try proxy editing.
CJB wrote on 12/1/2014, 9:16 PM
for some reason I get better playback with all intra AVCHD files (.mts container) than I do ProRes or DNxHD files.... so much for intermediates.....
johnmeyer wrote on 12/1/2014, 9:49 PM
for some reason I get better playback with all intra AVCHD files (.mts container) than I do ProRes or DNxHD files.... so much for intermediates.....That is a little unusual, but I have seen this on computers with relatively slow disk drives.

Since intermediates are 5-10 times larger, as I explained in my last post, they get large enough that a slow hard drive can become a bottleneck for fast playback. This would definitely be true if you are playing your video from external drives connected via USB 2.0.

There is no exact rule of thumb, and a lot depends on what settings you use to create your intermediate.

Even if your hard drive isn't a bottleneck when you are only reading one stream of video from that drive, it can get to be a problem if you have 3-4 streams of video on separate tracks (e.g., a multi-cam project), and the computer has to read all three streams from large intermediates that are all stored on the same drive.

john_dennis wrote on 12/2/2014, 12:40 AM
"[I] ...it can get to be a problem if you have 3-4 streams of video on separate tracks (e.g., a multi-cam project), and the computer has to read all three streams from large intermediates that are all stored on the same drive.[/I]"

See my actual measurements of this scenario in my March 17, 2014 post in this thread.

My main system:
Motherboard: Asus X99-AII
CPU: Intel i7-6850K
GPU: Sapphire Radeon RX480-8GB
RAM: Corsair Dominator (4 x 4 GB) DDR4 2400
Disk O/S & Programs: Intel SSD 750 (400 GB)
Disk Active Projects: 1TB & 2TB WD BLACK SN750 NVMe Internal PCI Express 3.0 x4 Solid State Drives
Disk Other: WD Ultrastar/Hitachi Hard Drives: WDBBUR0080BNC-WRSN, HGST HUH728080ALE600, 724040ALE640, HDS3020BLA642
Case: LIAN LI PC-90 Black Aluminum ATX Full Tower Case
CPU cooling: Corsair Hydro series H115i
Power supply: SeaSonic SS-750KM3 750W 80 PLUS GOLD Certified Full Modular Active PFC Power Supply
Drive Bay: Kingwin KF-256-BK 2.5" and 3.5" Trayless Hot Swap Rack with USB 3
Sound card: Crystal Sound 3 on motherboard. Recording done on another system.
Primary Monitor: Asus ProArt PA248q (24" 1920 x 1200)
O/S: Windows 10 Pro 190943
Camera: Sony RX10 Model IV

https://www.youtube.com/user/thedennischannel

johnmeyer wrote on 12/2/2014, 3:05 AM
I remember that thread. Those measurements were useful for illustrating the impact of file size on render speed, but I'm not sure they shed light on this issue because in this thread we're talking about timeline playback performance. However, I'm pretty sure one of us tested playback performance in one of the previous threads on this topic.
Paul Fierlinger wrote on 12/2/2014, 6:09 AM
John, you have opened my eyes and unclogged the bottleneck in my brain; I can't thank you enough. Since I work with different parameters than most or even everyone here I cannot always apply many of the thread's good advice to my profession but because you spent the time explaining the issue of intermediate files in unusual detail I found a way to apply this to my scenario.

In my line of work I draw every frame in an animation program called, TVPaint in HD 1080p quality but export to Vegas with the Motion-J-PEG codec as my intermediate format. I understand that this is not considered a cool format here, but my frames are a lot different than camera generated ones and Motion-JPEG is what works well for flat, hand drawn animation (wasn't the JPEG format actually invented for us?).

I have always exported my AVI files at 100% quality without thought; why go lower if I am already using an intermediate codec? (My final files are later all re-rendered anyway as uncompressed PNGs). But after reading your explanation, I lowered the Motion-JPEG value to 80% and to my great surprise the difference between 80 and 100% is barely recognizable, particularly when set to motion. My playback blues are now behind me -- and it was something so simple, sitting under my nose for years.