Quantcast ATI AIW VE observations, comparisons, etc - digitalFAQ Forum
  #1  
01-28-2019, 01:29 AM
Feedbucket Feedbucket is offline
Free Member
 
Join Date: Jun 2017
Location: NYC
Posts: 21
Thanked 3 Times in 3 Posts
So I did a few test captures with the All in Wonder VE I just installed and compared them to the same captures from my Hauppauge HVR-1600. Source is that MiscTestPatterns DVD played back through a DMR-ES15. Some interesting observations.

The good: The various histograms in AviSynth show that the signals on the VE are overall a lot more stable and closer to the source in shape than the HVR-1600, whose histograms are full of spikes meaning (as I understand it) banding and/or missing chunks of data. The 1600 also has some quirk where it somehow interlaces chroma (this was the issue I'd mentioned in an earlier post of mine) -- no such weirdness with the VE.

The bad: See the attached captures. Look closely at the blocks of solid color and you'll see that the VE is prone to dot crawl. This appears to be systematic rather than a result of interference/bad cables, as I tried two sources and two sets of cable and the result is the same. The VE also has some trouble accurately reproducing the 4:2:0 green box in the lower left of the S&W test pattern -- the alternating horizontal lines appear to be halved in height resulting in twice as many rendered in order to cover the same amount of space. The 1600 has neither of these issues.

I can appreciate that the VE probably provides a more "complete" picture through some proper analog-to-digital conversion + processing that the 1600 likely doesn't do, whereas the 1600 likely operates more like a photocopier, but based on eyesight alone the 1600's results seem to be a better 1:1 representation of the original sample. I know how highly regarded the early Theater chipset is around these parts, so the fact that the VE is apparently not quite living up to the hype is kind of making me wonder what's going on. Not sure if this is the expected behavior and hoping someone could contextualize some of what I'm seeing. Thanks.


Attached Files
File Type: avi testpatterns-20-dvd.avi (4.30 MB, 3 downloads)
File Type: avi testpatterns-20-hvr1600.avi (5.50 MB, 2 downloads)
File Type: avi testpatterns-20-aiwve.avi (5.64 MB, 4 downloads)
Reply With Quote
Someday, 12:01 PM
admin's Avatar
Ads / Sponsors
 
Join Date: ∞
Posts: 42
Thanks: ∞
Thanked 42 Times in 42 Posts
  #2  
01-28-2019, 01:38 AM
lordsmurf's Avatar
lordsmurf lordsmurf is offline
Site Staff | Video
 
Join Date: Dec 2002
Posts: 8,516
Thanked 1,397 Times in 1,227 Posts
I'm not sure how you came to that conclusion. I'm not seeing dot crawl, even viewed at 200%.

- Did my advice help you? Then become a Premium Member and support this site.
- Find television shows, cartoons, DVDs and Blu-ray releases at the TVPast forums.
Reply With Quote
  #3  
01-28-2019, 02:58 AM
Feedbucket Feedbucket is offline
Free Member
 
Join Date: Jun 2017
Location: NYC
Posts: 21
Thanked 3 Times in 3 Posts
Hmm, maybe I did something wrong when making those excerpts. Here are some screenshots: look at the pink/purple boxes on aiwve.png vs the other two. The checkerboard effect... is that not called dot crawl?

Also, the box with the lines inside it right above. They're thinner and twice as many in aiwve.png vs the HVR and source DVD. Something in the way ATI does colorspace conversion? The 600 USB does the same thing.

Edit: I just looked at the 600 USB again... it also has the checkerboard effect. Is this all ATI behavior in general?


Attached Images
File Type: png aiwve.png (67.0 KB, 14 downloads)
File Type: png hvr1600.png (52.9 KB, 10 downloads)
File Type: png dvd.png (46.1 KB, 10 downloads)
Reply With Quote
  #4  
01-28-2019, 03:34 PM
lordsmurf's Avatar
lordsmurf lordsmurf is offline
Site Staff | Video
 
Join Date: Dec 2002
Posts: 8,516
Thanked 1,397 Times in 1,227 Posts
I've seen that before, but it's caused by the signal and hardware in the chain. ATI cards, like all other hardware, can react differently (or badly) to other items in a workflow.

Dot crawl is almost always moving, not static, and more harsh. That is faint noise, and I don't think that's dot crawl.

I don't get this on my VE, using my normal hardware stack. But I have seen it when testing other hardware, not my main workflow gear. In fact, I took several sample capture with a Cypress TBC last year, where this was the issue. It wasn't seen at all in motion, only when stopping and analyzing frames at 200%+ size.

- Did my advice help you? Then become a Premium Member and support this site.
- Find television shows, cartoons, DVDs and Blu-ray releases at the TVPast forums.
Reply With Quote
The following users thank lordsmurf for this useful post: Feedbucket (01-28-2019)
Reply




Similar Threads
Thread Thread Starter Forum Replies Last Post
Using Panasonic ES15 after Mitsubishi HD2000U, observations stevevid Capture, Record, Transfer 3 08-25-2018 01:06 AM
VHS and S-VHS VCR comparisons (please contribute!) msgohan Project Planning, Workflows 4 05-23-2014 04:23 PM
Myth: Comparisons of Analog and Digital Video Resolution tomswift Tech Myths, Misinformation 39 06-19-2013 07:10 PM
Observations of using RAID-0 on a video editing computer manthing Computers 2 02-07-2012 05:12 AM
ATI TV Wonder HD 600 USB observations... NJRoadfan Capture, Record, Transfer 4 07-15-2011 12:23 AM

Thread Tools



 
All times are GMT -5. The time now is 03:29 AM