According to my tests of its DVD-Audio capabilities, when the Denon DVD-5900 is set for all small speakers with a subwoofer, the bass from the full-range channels is redirected to the sub. But when all of the main speakers are set to full range and the sub is turned off, the LFE information in the subwoofer channel is not redirected to the left and right front channels, as it should be. (The test disc used for DVD-A was The Ultimate DVD Surround Sam-pler & 5.1 Set-Up Disc, Chesky CHDVD221. I used the Bass Management Test to check for LFE bass redirection to the L/R front speakers, and the circular bass pan to test for the redirection of bass from the main speakers to the subwoofer.)

As measured from a preliminary Sony SACD test disc with the Denon set to perform bass management on SACDs, I confirmed that the Denon's bass management rolls off the low frequencies in the L/R speakers when they are set to Small and redirects it to the subwoofer (the test disc is 2-channel only).

The chroma bug was not evident from the DVD-5900's interlaced output into a non-DCDi deinterlacer (DCDi tends to hide the chroma bug). Using the test patterns on Avia PRO, a layer change on the Denon took just over three seconds. (This duration includes the reaction time that's unavoidable with a manual stopwatch, but was approximately the same on 15 attempts.) I observed a sync error (video delay) of three frames from both the component progressive and DVI outputs.

The DVD-5900's luminance (black and white) response measured –0.92dB at 5MHz and –1.5dB at 5.5MHz in interlaced component (–0.35 and –0.92, respectively, in progressive component), and remained visible on test patterns out to the 6.75MHz limit of DVD from both the component and DVI outputs. The red-and-blue color bandwidth began to roll off above about 2.5MHz, and, for all practical purposes, the color information was gone by the 3.38MHz limit of the DVD format. (On a DVD, the color is limited to half the bandwidth of the luminance.) This was true from both the progressive and interlaced component outputs, and was likely caused by the color-bandwidth limiting that's an inherent and intentional characteristic of Faroudja processing. (Both the Denon player and the Marantz projector use Faroudja DCDi chips.) The color also started rolling off above about 2.5MHz from the DVI output. In this case, however, it was still easily visible on the 3.38MHz multiburst test pattern, though clearly well down in level.

The deinterlacing of the DVI and progressive component outputs was virtually flawless in the motion tests on the Faroudja test DVD, and the Denon DVD-5900 captured an unflagged 3:2 pulldown sequence as quickly as any progressive player I've tested.—TJN