Raritan, Inc. - The Tolly Group

T
H
E
TOLLY
G R O U P
No. 206157
October 2006
Raritan, Inc.
Test
Summary
®
Paragon II Cat5 KVM Switch & Enhanced
User Station Video Quality Evaluation
versus Avocent® AMX Switch and Enhanced User Station
aritan commissioned The Tolly
Group to evaluate the video quality of its enterprise-class Paragon® II Cat5 KVM Switch and
Enhanced User Station analog Keyboard/
Video/Mouse (KVM) solution against
Avocent’s AMX® solution. Both
solutions provide analog KVM over unshielded twisted pair (UTP) cables. For
this evaluation, we used switches that can
support up to four users and 42 servers.
R
Tests involved the subjective evaluation of
the video quality of both solutions at various display resolutions in terms of video
bandwidth, color gradient, jitter, distortion
and ghosting/reflection, using various
cable lengths. A video expert rated the
solutions during a series of blind tests,
with brand identities hidden.
Raritan’s Paragon II KVM solution
delivered an overall better video quality
than Avocent’s solution tested. Tests were
conducted in October 2006.
© 2006 The Tolly Group
Test Highlights
Delivers an overall better video quality measured in terms of
video bandwidth, color gradients, jitter, distortion and
ghosting compared to the Avocent solution tested
Maintains superior video quality over the shortest and
longest cable length and highest resolution compared to the
Avocent solution
Provides an option for manual and automatic skew
correction for longer length cabling
Composite Video Quality Score*
Raritan Paragon II versus Avocent AMX
Composite video quality score
Premise: With the increasing use of videointensive applications (i.e., CAD/CAM, and
broadcast applications), end users require a
KVM solution that can handle extremely
high video resolutions. In addition, since
many of the end-users of these high resolution video applications are outside the data
center, IT departments need to deliver highresolution access to multiple servers over
greater distances. For end-users whose
work requires them to spend hours viewing
high-resolution graphics and/or video, high
quality video is essential. Poor video quality
cannot only impact accuracy and productivity, but it can also be detrimental to the
vision of end users.
4.0
3.5
Raritan Paragon II
Avocent AMX
3.5
3.2
3.0
3.1
2.9
2.9
2.5
2.6
2.7
2.0
1.5
1.3
1.0
0.5
0.0
100 ft
300 ft
600 ft
1,000 ft
Cable length (Feet)
Composite video quality score of:
4 = Good, crisp video image with no apparent distortion
3 = Good, crisp video image with minimal distortion
2 = Fair and viewable image with noticeable distortion
1 = Poor image with too much distortion
Note: Video quality was measured by an independent video expert in a blind test See the
Executive Summary section of the report for details.
Source: The Tolly Group,October 2006
Figure 1
Page 1
Raritan, Inc.
ParagonⓇ II Cat5 KVM Switch and Enhanced User Station
Executive
Summary
Raritan’s Paragon II
Cat5 analog KVM
solution delivers and
maintains superior
video quality compared
to Avocent’s AMX
solution tested.
The Tolly Group’s evaluation
focused on a solution involving a Raritan Paragon II Cat5
UMT442 KVM switch and a
P2-EUST Enhanced User Station. The Paragon II solution
was tested against an Avocent
AMX5020 KVM switch and
an AMX5130 user station.
Each KVM solution was
© 2006 The Tolly Group
evaluated for its video bandwidth, color gradients/
sharpness, jitter and distortion,
and finally, ghosting/reflection.
Though the quality of video delivered by a KVM solution is
subjective by nature, The Tolly
Group took a number of steps to
ensure a controlled test environment. The company exposed
both KVM solutions to the same
tests over cable lengths of 100,
300, 600 and 1,000 feet.
Further, the company invited a
subject matter expert in the area
of digital audio-visual communications systems to act as an
impartial evaluator of the video
quality. The expert, a professor
at a state university in Florida,
holds a PhD in Electrical Engineering from Columbia University and regularly contributes to
the MPEG-4 video standard.
Tests were conducted where the
video expert viewed test patterns delivered by the KVM solutions tested and assessed the
quality of the two solutions —
although the product brands
being tested were not revealed
to the subject expert. Furthermore, the tests were repeated
with products swapped between
the display monitors unbeknownst to the video expert, to
ensure a fair test.
The video expert evaluated
image quality and assigned a
subjective score on a scale of 1
to 4 (4 = Good, crisp video with
no apparent distortion, 3 =
Good, crisp video image with
minimal distortion, 2 = Fair and
viewable image with noticeable
distortion, and 1 = Poor image
with too much distortion).
Page 2
Raritan, Inc.
ParagonⓇ II Cat5 KVM Switch and Enhanced User Station
In every test, in almost all the
areas tested, the Raritan Paragon II switch demonstrated
superior video quality over the
Avocent solution tested. In
fact, across the four major test
scenarios, the Raritan Paragon
II switch matched or clearly
exceeded the performance of
the Avocent AMX solution.
(A comparison of the composite video scores for each solution under test (SUT) for the
various cable lengths tested is
shown in Figure 1, and individual video quality scores for
each SUT and test scenario is
shown in Figure 2.)
Video bandwidth
This test measured the sharpness and horizontal resolution
of the displayed video using
the “Misconvergence 2” test
pattern from the NEC Monitor
Test application, see Figure 3.
The test verified if the SUT
provided enough end-to-end
horizontal and vertical video
bandwidth from the video output of the server all the way to
the end user station, to faithfully preserve the video quality. Lack of sufficient video
bandwidth results in the loss
of horizontal or vertical lines
in the test pattern. The displayed video image was examined for sharpness and any
loss of vertical lines, which
indicates lack of horizontal
resolution bandwidth.
Test results showed that
Raritan’s Paragon II Cat5
KVM solution achieved consistently better video quality
scores than Avocent’s AMX
KVM solution as tested across
all the cable lengths, display
resolutions and refresh rates.
The Raritan solution achieved
an average score of 3.08 versus
© 2006 The Tolly Group
2.43 for the Avocent solution
across all cable lengths.
Raritan, Inc.
Color Gradients &
sharpness
Paragon II
Cat5 KVM
This test evaluated the ability
of the KVM solution to
maintain uniform color
intensity and contrast over
different cable lengths and
display resolutions.
Competitive
Video Quality
Evaluation
versus Avocent AMX
Tests once again showed that
Paragon II Cat5 KVM solution achieved better video
quality scores than Avocent’s
AMX solution, for 100 ft and
300 ft cable lengths for all
display resolution settings.
Avocent scored slightly better
than Raritan for just the 600 ft
cable length, while Raritan
clearly outperformed Avocent
at 1,000-ft length. On average, Raritan scored 2.88 versus 2.4 for the Avocent product across all cable lengths.
Tests used the “Color Bars”
test pattern from the NEC
monitor test. (See Figure 3.)
Jitter & Distortion
This test evaluated the ability
of the KVM solution to present a jitter and distortion-free
image over varying cable
lengths and display resolutions. The “Test Picture”
pattern from the NEC monitor
test application was used for
this test. See Figure 3.
Once again, test results
showed that the Raritan Paragon II achieved better video
quality scores than Avocent
— 3.0 for Raritan versus 2.5
for Avocent. Raritan and Avocent were evenly matched at
cable lengths of 100 and 300
ft, while Avocent performed
slightly better than the Raritan
Paragon II solution at 600 ft.
Product Specifications
Vendor-supplied information not
necessarily verified by The Tolly Group
Raritan, Inc.
Paragon II Cat5 KVM Switch
Industry-leading Cat5 analog KVM
switch with ultra-high performance
video
Plug-and-play auto-configured
installation
“Keep-alive” technology for continuous server availability
Industry’s highest video resolution: Up
to 1920 x 1440 at distances of 1,000
ft. at a refresh rate of 60Hz
Cross-platform support for PS/2,
USB, Sun and serial devices
Automatic and manual skew compensation with the P2-EUST for video
applications with extremely stringent
signal tolerances
Automatic and manual video tuning
with the P2-EUST, video tuning
occurs automatically, saving time on
server setup, or it can be manually
adjusted
Industry’s broadest range of port densities — non-blocked access up to 64
users have a clear path to up to
10,000 servers
Optional UST-IP user station provides
KVM-over-IP access and control of
servers
For more information contact:
Raritan, Inc.
400 Cottontail Lane
Somerset, NJ 08873
Ph: (800) 724-8090
Fax: (732) 764-8887
URL: http://www.raritan.com/4381
Page 3
Raritan, Inc.
ParagonⓇ II Cat5 KVM Switch and Enhanced User Station
However, at 1,000 ft, Raritan
clearly outperformed Avocent
across all display resolutions
and refresh rates.
Ghosting or
Reflection
Engineers also measured the
ability of the KVM solutions
under test to preserve the quality of text of different font
sizes, (See Figure 3) over varying cable lengths and display
resolutions.
Displayed video was examined for the presence of ghosting (hazy outline of the text
shapes around the letters) or
reflection (hazy outline of the
text shapes inside the letters).
The tests once again showed
that the Paragon II delivered
better video quality by preventing ghosting or reflection of the
text across all scenarios.
The Raritan Paragon II achieved
an average score across all cable
lengths of 3.28 versus 2.63 for
the Avocent product.
Test bed setup &
methodology
The solutions tested consisted of
Raritan’s Paragon II solution and
Avocent’s AMX KVM solution.
The test infrastructure consisted
of two identical servers and CRT
monitors each connected to one
SUT. Each server was once again
configured with identical video
cards. (See Figure 4.)
The Paragon II solution consisted
of a UMT442 KVM switch, a
P2-EUST Enhanced User Station
and a P2-CIM (Computer Interface Module) which essentially
converts the standard video out-
put of the keyboard, video and
mouse outputs of the server to a
UTP (RJ-45) port.
The comparable SUT from
Avocent consisted of an
AMX5020 KVM switch, an
AMX5130 desktop user station,
and an AMIQ-PS2 Server Interface Module, which converts the
keyboard, video and mouse outputs of the server into a
single UTP (RJ-45) port.
For each SUT, the server was
connected to the KVM switch
using the supplied interface
module adapter and a (Cat 5e)
UTP cable. The user port on the
KVM switch was then connected to the corresponding user
station using a Belden CDT
Networking DataTwist 1700
series UTP (Cat 5e) cable of desired length (100-, 300-, 600and 1,000-ft lengths were
TEST PATTERNS
© 2006 The Tolly Group
Page 4
Raritan, Inc.
ParagonⓇ II Cat5 KVM Switch and Enhanced User Station
tested). The user station was
then connected to a 21” ViewSonic G225fB CRT monitor
using a VGA cable. The CRT
monitors were capable of displaying 1920x1440 resolution
at 32-bit color depth and 60Hz
refresh rate.
A publicly available NEC
Monitor Test application was
used for evaluating the video
quality of each KVM solution
over different cable lengths.
See Figure 3 for the test patterns used from this test application. In addition to the NEC
Monitor Test application, a
Microsoft Word document was
created with variable text sizes
to examine how each product
preserved the quality across
different cable lengths and
display resolutions.
Test scenarios
Tests were run at video resolutions of 1280x1024 (75Hz and
85Hz refresh rates),
1600x1200 (75Hz and 85Hz
refresh rates) and 1920x1440
(60Hz refresh rate). For each
test, the SUTs used identical
cables of the desired length.
The tests were run by Tolly
Group engineers and the video
quality of the displayed
images was evaluated by the
independent video expert in a
blind test. During the test, the
engineers switched the displays between the two SUTs
unbeknownst to the video expert. The tests were run a second time with the monitors
connected to a different SUT
than the first run, in order to
ensure a fair test.
At the completion of testing,
Tolly Group engineers calculated a composite score for
each SUT, for each tested
cable length. For each cable
length, there were 20 test
© 2006 The Tolly Group
scenarios, and the video quality
score achieved in all the test
scenarios for a particular cable
length were averaged.
Video Bandwidth Test
For this test, engineers used the
“Misconvergence 2” test pattern
from the NEC monitor test application. The displayed image
was checked for sharpness and
horizontal resolution bandwidth.
The vertical lines in the displayed image had to be visible
and distinct, the lack of which
indicates bandwidth issues that
affect horizontal resolution.
The video expert assigned a
video quality score for each
SUT under the various test scenarios described earlier.
Color gradients test
For this test, engineers used the
“Color Bars” test pattern. Under
the various test scenarios mentioned earlier, displayed video
was evaluated for each SUT.
The image was checked for
color gradient changes and contrast, with the requirement that
there be no visible distortion or
smearing of the color bars on to
the black background.
Jitter/Distortion test
In this test, engineers used the
“Test Picture” test pattern (See
Figure 3). For all the various test
scenarios outlined earlier, the
video expert evaluated the video
quality of each SUT.
The requirement for passing the
test mandated that all the distinct grids in the crosshatch pattern be clearly visible, and be of
equal brightness. Vertical lines,
circles and text info in the test
pattern had to be clearly visible.
The presence of jitter could be
identified in the form of “sparkling pixels” on the black
background.
Ghosting or
Reflection test
For this test, the engineers used
a Microsoft Word document
(See Figure 3). The Word
document contained text of
different font sizes, and boldness levels.
Fair Testing Charter™
Interaction with Competitors
In accordance with The Tolly Group’s process, competitors were contacted and invited to review the test
plans, the product levels and configurations of their
products and to review and comment on their results.
Avocent accepted this invitation. For more information
on this process, please see: http://www.Tolly.com/FTC.aspx.
Avocent representatives attended the product evaluation and were given
the opportunity to verify the configuration of the Avocent equipment and
the test environment. Following a review of the results, Avocent said,
“The Avocent AMX switching system was tested using only the automatic video equalization and de-skew settings. The AMX5130 user station allows manual fine-tuning of both the video equalization and video
de-skew settings to optimize the video quality at the user station. The
manual fine-tune settings are most beneficial when accessing higher
resolution target devices at longer cable distances, such as the 600 foot
and 1,000 foot distances that were tested by The Tolly Group.”
Page 5
ParagonⓇ II Cat5 KVM switch and Enhanced User Station
Raritan, Inc.
The requirement of the test
was that the text should be
clear and free of ghosting
or reflections around or
inside the text. The video expert
evaluated the quality of the displayed video of each SUT under
each of the test scenarios listed
earlier.
Test Bed Diagram
ViewSonic G225fB
21" PerfectFlat CRT
Monitor
VGA, Keyboard
and Mouse
Connectors
ViewSonic G225fB
21" PerfectFlat CRT
Monitor
`
`
The Tolly Group is a leading global provider of thirdparty validation services for
vendors of
IT products,
components and
services.
VGA, Keyboard
and Mouse
Connectors
Avocent AMX5130
Desktop User Station
Raritan Paragon II
P2-EUST
Enhanced User Station
UTP cable
UTP cable
Avocent
AMX5020
KVM Switch
Raritan Paragon II
UMT442 KVM
Switch
UTP Cable
connected to
Raritan P2CIM
The company is based in
Boca Raton, FL and can
be reached by phone at
(561) 391-5610, or via the
Internet at
http://www.tolly.com,
sales@tolly.com
UTP Cable connected to
Avocent AMIQDM Server
Interface Module
Server with 3-GHz Intel Pentium 4
HT CPU, 1GB DDR2 RAM and ATI
Radeon X1300 Pro Graphics Card
Server with 3-GHz Intel Pentium 4
HT CPU, 1GB DDR2 RAM and ATI
Radeon X1300 Pro Graphics Card
Test Equipment/Software Summary
The Tolly Group gratefully acknowledges the providers of test equipment/software used in this project.
Vendor
Product
Web
NEC Deutschland GmbH
NEC Monitor Test Program
http://www.benchmarkhq.ru/engl
ish.html?/be_monitor.html
(available in public domain)
206157-sbwfm1-cdb-12oct06
© 2006 The Tolly Group
Page 6
Download PDF
Similar pages