13 Replies Latest reply on Dec 10, 2010 7:58 PM by Butch2oc

    Bit Depth and Render Quality

    Butch2oc Level 1

      When you finally export media to some sort of media format via the encoder does the projects preview Bit Depth and Render Quality settings affect the output file?

      I know there is "Use Preview files" setting in the media exporter dialogue but I just want to be sure of what I am doing.

        • 1. Re: Bit Depth and Render Quality
          Harm Millaard Level 7

          Use preview files is best not used on export.

           

          Use Maximum Render Quality improves the final quality but also increases the encoding time significantly.

           

          Use Maximum Bit Depth is only useful with AJA/BM cards, and ingest over HD-SDI in 4:2:2 format. Otherwise it will only increase storage requirements, slow down the encoding process and just pad the color info with zeroes.

          • 2. Re: Bit Depth and Render Quality
            Colin Brougham Level 6

            Use Maximum Bit Depth is only useful with AJA/BM cards, and ingest over HD-SDI in 4:2:2 format. Otherwise it will only increase storage requirements, slow down the encoding process and just pad the color info with zeroes.

             

            This is a false statement. Maximum Bit Depth forces the renderer to work in 32-bit mode, so that 32-bit effects are processed with a greater degree of accuracy, regardless of the export module. A number of effects in Premiere Pro CS5 are 32-bit; use the "32" filter in the Effects Panel to reveal those. Elements like blurs, glows and shadows are processed differently with MBD, and have a different effect on the output.

             

            View the following thread for some pictorial examples of what Maximum Bit Depth does: Battling Hardware MPE, Episode 2: Chunky Blurs

            • 3. Re: Bit Depth and Render Quality
              Harm Millaard Level 7

              Colin,

               

              Please explain to me when you start with 4:2:0 compressed material, where the extra information is coming from to use 4:2:2 format, other than padding with zeroes? Just like starting with Xvid and converting to uncompressed does not add information, that has been thrown away in the compression stage, so where is that added information coming from?

               

              If you statement were true, then

               

              a. there is no need for expensive 4:4:4 of less costly 4:2:2 cameras, because by magic the lost info reappears.

               

              b. we can all just add a number of zeroes to our 4:2:0 material and like magic we have the same quality as 4:4:4 cameras now have.

               

              I don't follow that argument.

              • 4. Re: Bit Depth and Render Quality
                Jeff Bellune Adobe Community Professional

                Colin's probably working on a response, but in the meantime I'll add my perspective.

                 

                The first rule of editing should be, "Do no harm".  That's why we have lossless codecs for DIs.  By moving the video into 32-bit space and processing 32-bit effects in 32-bit space, then returning the video to its original bit depth for export, we have theoretically preserved whatever quality was in the original without degrading it further by effects processing.  That's why audio is conformed to 32-bit on import, and why Adobe Audition allows 32-bit processing before exporting 16-bit CD audio.

                 

                It's not magic, it's preservation.

                 

                -Jeff

                • 5. Re: Bit Depth and Render Quality
                  Colin Brougham Level 6

                  Jeff's response is my perspective, as well, which is both backed up by my own tests and the official Adobe word.

                   

                  Exhibit A: My Tests

                   

                  EXPORT-BLUR-TESTS.png

                   

                  That is DV footage with a title superimposed over it in a DV sequence, with a Gaussian blur effect (the Premiere accelerated one) applied to the title; all samples are from that sequence exported back to DV. This was to show the relative differences of processing between software and hardware MPE, Premiere export and AME queueing, and the effect of the Maximum Bit Depth and Maximum Render Quality options on export (not the sequence settings; those have no bearing on export).

                   

                  The "blooming" evident in the GPU exports is due to hardware MPE's linear color processing. I think it's ugly, but that's not the point here. Further down the line, you can see the effect of Maximum Bit Depth (and MRQ) on both software MPE and hardware MPE. I assume you can see the difference between the Maximum Bit Depth-enabled export and the one without. Bear in mind that this is 8-bit DV footage composited and "effected" and exported back to 8-bit DV. I don't understand what your "padding with zeroes" and larger file size argument is motivated by--my source files and destination files are the same size due to the DV codec--but it's plainly clear that Maximum Bit Depth has a significant impact on output quality. Similar results would likely be evident if I used any of the other 32-bit enabled effects; many of the color correction filters are 32-bit, and should exhibit less banding, even on something 8-bit like DV.

                   

                  Exhibit B: The Adobe Word

                   

                  This is extracted from Karl Soule's blog post, Understanding Color Processing: 8-bit, 10-bit, 32-bit, and more. This section comes from Adobe engineer Steve Hoeg:

                   

                  1. A DV file with a blur and a color corrector exported to DV without the max bit depth flag. We will import the 8-bit DV file, apply the blur to get an 8-bit frame, apply the color corrector to the 8-bit frame to get another 8-bit frame, then write DV at 8-bit.

                   

                  2. A DV file with a blur and a color corrector exported to DV with the max bit depth flag. We will import the 8-bit DV file, apply the blur to get an 32-bit frame, apply the color corrector to the 32-bit frame to get another 32-bit frame, then write DV at 8-bit. The color corrector working on the 32-bit blurred frame will be higher quality then the previous example.

                   

                  3. A DV file with a blur and a color corrector exported to DPX with the max bit depth flag. We will import the 8-bit DV file, apply the blur to get an 32-bit frame, apply the color corrector to the 32-bit frame to get another 32-bit frame, then write DPX at 10-bit. This will be still higher quality because the final output format supports greater precision.

                   

                  4. A DPX file with a blur and a color corrector exported to DPX without the max bit depth flag. We will clamp 10-bit DPX file to 8-bits, apply the blur to get an 8-bit frame, apply the color corrector to the 8-bit frame to get another 8-bit frame, then write 10-bit DPX from 8-bit data.

                   

                  5. A DPX file with a blur and a color corrector exported to DPX with the max bit depth flag. We will import the 10-bit DPX file, apply the blur to get an 32-bit frame, apply the color corrector to the 32-bit frame to get another 32-bit frame, then write DPX at 10-bit. This will retain full precision through the whole pipeline.


                   

                   

                   

                  6. A title with a gradient and a blur on a 8-bit monitor. This will display in 8-bit, may show banding.

                   

                  7. A title with a gradient and a blur on a 10-bit monitor (with hardware acceleration enabled.) This will render the blur in 32-bit, then display at 10-bit. The gradient should be smooth.

                   

                  Bullet #2 is pretty much what my tests reveal.

                   

                  I think the Premiere Pro Help Docs get this wrong, however:

                   

                  High-bit-depth effects

                  Premiere Pro includes some video effects and transitions that support high-bit-depth processing. When applied to high-bit-depth assets, such as v210-format video and 16-bit-per-channel (bpc) Photoshop files, these effects can be rendered with 32bpc pixels. The result is better color resolution and smoother color gradients with these assets than would be possible with the earlier standard 8 bit per channel pixels. A 32-bpc badge appears to the right of the effect name in the Effects panel for each high-bit-depth effect.

                   

                  I added the emphasis; it should be obvious after my tests and the quote from Steve Hoeg that this is clearly not the case. These 32-bit effects can be added to 8-bit assets, and if the Maximum Bit Depth flag is checked on export, those 32-bit effects are processed as 32-bit, regardless of the destination format of the export. Rendering and export/compression are two different processes altogether, and that's why using the Maximum Bit Depth option has far more impact than "padding with zeroes." You've made this claim repeatedly, and I believe it to be false.

                   

                  Your witness...

                  • 6. Re: Bit Depth and Render Quality
                    Harm Millaard Level 7

                    Colin,

                     

                    Sorry to be so obstinate, but bear vith me. I have a 8 bit DV clip. Standard stuff.

                    I do not apply effects or transitions. I export to standard DV AVI, again 8 bits.

                     

                    Where is the benefit of maximum bit depth? It pads color space with zeroes and removes them again.

                     

                    If you have eight candies in a box and you add two empty wrappers to that box, there will still be only eight candies, but you now have 10 wrappers in that box. You do your stuff like adding effects or transitions with maximum bit depth, and at the end you remove two wrappers on export, you hope that those will be the empty wrappers, because if you choose the wrong wrappers, you may end up with 6 or 7 candies.

                    • 7. Re: Bit Depth and Render Quality
                      Jeff Bellune Adobe Community Professional

                      I do not apply effects or transitions

                      In that specific case, then you are correct.  There is no benefit to exporting at max bit depth.  There's also no benefit if none of the effects or transitions you do use are 32-bit.  I believe Colin was very clear on both points.

                       

                      -Jeff

                       

                      EDIT @ Colin below: Jinx!  Do I owe you a beverage of your choice now?

                      • 8. Re: Bit Depth and Render Quality
                        Colin Brougham Level 6
                        Sorry to be so obstinate, but bear vith me.

                         

                        Harm, I would expect nothing less

                         

                        I have a 8 bit DV clip. Standard stuff.

                        I do not apply effects or transitions. I export to standard DV AVI, again 8 bits.

                         

                        Where is the benefit of maximum bit depth? It pads color space with zeroes and removes them again.

                         

                        Correct; in that scenario, there is no benefit. This is indicated in the quotes from Karl's blog post. However, my retort was based on your claim of Maximum Bit Depth being "only useful with AJA/BM cards, and ingest over HD-SDI in 4:2:2 format." In the examples I posted, graphical and otherwise, this is clearly not the case. Even DV footage can benefit from the use of Maximum Bit Depth when specific effects are employed.

                         

                        Your point is not wrong; your qualifier of "is only" is wrong, however.

                         

                        EDIT: @ Jeff: Jinx!

                        • 9. Re: Bit Depth and Render Quality
                          Colin Brougham Level 6

                          EDIT @ Colin below: Jinx!  Do I owe you a beverage of your choice now?


                          Several, I do believe...

                           

                          (Shh... the Lounge Police are coming!)

                          • 10. Re: Bit Depth and Render Quality
                            Jim_Simon Level 8
                            It pads color space with zeroes and removes them again.

                             

                            Maybe it's me, but I have a different understanding of the process.  Moving from 8 bit to 32 bit color is not the same as going from a 4:2:0 color subsampling to 4:2:2 subsampling.  The bit depth affects the degrees of color possible in any given channel, but not how often those channels are sampled.  Bit depth relates more to accuracy, and subsampling more to quantity.  Two different things.

                             

                            To continue your candy analogy, 32 bit color is like slicing those same 8 candys into a total of 32 pieces, whereas changing the subsampling would be adding extra boxes.  Whether those boxes contain actual candy or just wrappers would depend on when the boxes came into existence, before recording or after.  But at no point are extra wrappers added to any preexisting box.

                            • 11. Re: Bit Depth and Render Quality
                              Butch2oc Level 1

                              Interesting discussion guys... I'm just a little more unsure now

                               

                              So... I'm working on a 10bit YUV Uncompressed BM timeline.

                              I uncheck Maximum Bit Depth and Render Quality so my general editing is a little quicker.

                              I then render final timeline for viewing with the settings still unchecked.

                               

                              what bit depth and quality is the rendered file at this stage?

                               

                              from your debate above I'm guessing that the rendered file is still 10 bit but the 32bit plugs used on timeline are not rendered with precision.

                               

                              ...The client wants an approval file now... so... I export media utilising "use preview files" as they only want an mpeg to view.

                               

                              ...Client now approves file and I can export a final (best possible quality file) for transmission.

                               

                              So I now export with Maximum Bit Depth and Render Quality checked on the media encoder panel.

                               

                              do these settings in the encoder over ride the Maximum Bit Depth and Render Quality settings in PP?

                               

                               

                              Is this the correct workflow?

                              Do I need a box of candy?

                              • 12. Re: Bit Depth and Render Quality
                                Colin Brougham Level 6

                                I uncheck Maximum Bit Depth and Render Quality so my general editing is a little quicker.

                                 

                                That shouldn't be the case; those flags in the Sequence Settings only affect preview files, and those flags in the Export Settings affect output. They are mutually exclusive and not connected; one does not enable the other. These are RENDER settings, not playback settings, remember...

                                 

                                what bit depth and quality is the rendered file at this stage?

                                 

                                from your debate above I'm guessing that the rendered file is still 10 bit but the 32bit plugs used on timeline are not rendered with precision.

                                 

                                It depends. What is your preview codec set to for your sequence? If it's 10-bit YUV, then yes--it's 10-bit. If you're using something like DV or MJPEG, it's going to be 8-bit. If you've generated preview files, you've "cooked" them to whatever codec is set for the sequence, and the bit depth is dependent on the codec you've selected. Regardless of this, since you don't have Maximum Bit Depth selected in the Sequence Settings, all effects were processed as 8-bit. So even if your source footage is 10-bit, and your preview codec is 10-bit, you're incurring a loss of color bit depth without MBD set, since the effects are being processed in 8-bit.

                                 

                                ...Client now approves file and I can export a final (best possible quality file) for transmission.

                                 

                                So I now export with Maximum Bit Depth and Render Quality checked on the media encoder panel.

                                 

                                do these settings in the encoder over ride the Maximum Bit Depth and Render Quality settings in PP?


                                 

                                Only if you don't check the Use Previews option like you did for the client preview. If you do leave that checked, export will use the preview files that went from 10-bit source to 8-bit effects processing to 10-bit preview file. Maximum Bit Depth isn't really going to have any effect here, beyond maybe causing the export to take longer, since your effects are already cooked into your preview files... and you know what those have been through.

                                 

                                However, if you make sure to clear the Use Previews checkbox, and you set Maximum Bit Depth and Maximum Render Quality in the Export Settings, your sequence will pass a 10-bit source frame for the effects to be rendered in 32-bit color, before being compressed to your final destination file, and whatever color bit depth it uses.

                                 

                                Is this the correct workflow?

                                 

                                You decide...

                                 

                                Do I need a box of candy?

                                 

                                Undoubtedly. Or, re-read (or read it for the first time) Karl's blog post on this issue: The Video Road – Understanding Color Processing: 8-bit, 10-bit, 32-bit, and more

                                • 13. Re: Bit Depth and Render Quality
                                  Butch2oc Level 1

                                  Thanks Colin... All clear