Test 2: Saving Private Ryan
This movie is known to be notoriously hard to encode and therefore ideally fits my needs. I chose 3 scenes here: The beach landing (frames 8430 - 22281), a still scene in the controller's office (43790 - 45189) that has a lot of film grain and a night scene (frames 94122 - 106098). This should prove sufficiently difficult to get any codec into trouble. Let's see if I succeeded.
I don't think there's too much to be said about the beach scene. No steadycam, underwater shots, a lot of debris and confusion on the screen. Before the shooting starts there are many close-ups in the boat. As the boat door opens we can notice an immediate drop in details in all codecs. However, this effect is more visible in some codecs (3ivX, DivX3, NeroDigital) than in others. DivX3 also tends to smear the scene where the GIs are shot while still being in the boat.
The next scene shows the water from above.
Water is really not the element of MPEG-4 codecs. Let's get on shore..
There are a number of close-ups before the next scene. Those are pretty detailed (exception: ND is somewhat washed out), but with varying amounts of blocks (3ivX, DivX3 are pretty blocky).
The 2nd scene I reviewed is rather short and does not contain a great deal of movement. I therefore focused on faces and textures. As the screenshots are pretty nondescript, I have decided to drop them. All codecs did a pretty good job, but 3ivX, DivX3 and DivX5 sometimes had trouble to make a clear difference between the different elements of the window shutters.
The last scene is a night scene where our soldiers rest for the night.
The dark walls were not as problematic as they used to be. Only DivX3 had visible trouble and RV9 did the best job here.
As Tom Hanks goes to talk with his corporal (the guy who ask him where he's
from), we see Tom Hanks in front of a doorframe. RV9 had no problems with that
but the other codecs did. 3ivX turned out a really ugly doorframe with lots
of blocks, DivX3 had a lot of blocks and smearing, DivX5 was somewhat blocky,
NeroDigital had few blocks, VP6 was about comparable to DivX5 and so was XviD.
Last but not least a scene without comments.
I noticed a general improvement in the night scenes as compared to the last comparison (but not quite as much as the last time).
As in the Matrix test, XviD performed really well, keeping details while not being blocky. VP6 and RV9 also did rather well (the latter still exhibiting a lack of detail during motion scenes). As with Matrix, the codecs could be divided into the same two groups. Though I felt that RV9 handled Matrix somewhat better (less visible lack of details). The other codecs performed roughly the same.
Now you can proceed to the Futurama test or go directly to the conclusion.
This document was last updated on December 31, 2003