ABIT KV7 Motherboard Review

Supplied by ABIT

ABIT continues to take bold design step going places others don’t. The AT7 series broke new ground in full featured motherboards and now ABIT steps out on a limb with the KV7 KT600 motherboard. The KT600 chipset promises many improvements including reduced latency through faster access to system memory, improved six-channel sound and native serial ATA/RAID support. These improvements may be more of a benefit to the manufacturer than the enthusiast as many are discovering.

The KT600 chipset, as with all new chipsets, was thought to best the NFORCE2 platform. It didn’t. We’ve had only two experiences with KT600 motherboards and they have been anything but picture perfect compared to other models. Perhaps the driving reason behind VIA’s KT600 chipset is a demand from manufacturers to do more with less. The KT600 may be the answer to a platform that is cheaper to implement thus the manufacturer can stay competitive in the mainstream market. The truth is that most consumers, especially system builders, do not buy the higher priced enthusiast motherboards due to cost factors.

Specifications

Look inside the box

The ABIT KV7 is a no frills package designed to meet the needs of the cost conscious consumer. Inside there is the motherboard and the bare minimum of necessary hardware/software:

Manual

Install CD

Floppy SATA Raid drivers

1 EIDE cable and 1 floppy cable

2 x SATA cables

1 x SATA power cable

1 x USB cable

and backplane guard

We believe the reason for this is one of cost. The minimalist approach allows ABIT to produce a product at a very cost competitive price point that the system builder or consumer will like. The system builder for a company or retailer can build a computer that meets the performance needs of today with a lot of the required options. The KV7 would make for a very cost effective workstation for a business.

Touring the board

Right out of the gate the KV7 is a narrower ATX motherboard. The standard width is approximately 9.5 inches and the KV7 measures up at only 7 inches wide.

The KV7 delivers a lot in a smaller package but smaller isn’t always necessarily better. The socket area is quite crowded. The north end of the socket is surrounded by capacitors that may present a problem for attaching larger base heatsinks. Those capacitors will most definitely create an access impediment to mounting clips. It is a tight space. Most likely a screwdriver will have to be used to aid in attaching heatsink clips. ABIT has not forgotten to include four heatsink mounting holes around the socket.

With smaller spaces comes design challenges. The first oddity is the placement of the heatsink fan header. The header is placed further away from the socket than in most motherboard designs. It also looks like the ram lifters could touch the an installed fan header but they do not.

Three DIMM sockets are worth noting. It has been our experience that RAM should be placed in the DIMMS starting with the slot nearest the CPU. This is true for one module but when two are used there were stability problems. DIMM 2 and 3 should be used when memory is placed in pairs. Memory in DIMM 1 and 2 did cause stability issues.

At the south end of the DIMM slots the lifters are tight to the AGP slot but do not present a problem.

The video card lifter is a problem. It is nestled between four capacitors making for a tight squeeze to push the video card lifter. Those with fingers the size of baseball mitts will soon grow to dislike the two lower capacitors.

The northbridge fan is a debatable design feature. On the one hand it is larger and therefore looks like it does a more effective cooling job.

Two screws hold the top cover in place allowing for the fan to be lifted out.

On the other hand it is a proprietary piece of hardware compared to chipset coolers which feature a 40mm or even a 60mm. box fan. Those fans can be found at electronics stores if ever the northbridge fan should fail. If the ABIT fan fails then it is RMA time. The real reason for this style of fan may lie in marketing appeal and cost. The fan looks like it can do a more effective job and may be at a lower cost per unit than traditional box fan and chipset heatsink.

ABIT does use a generous amount of thermal paste between the heatsink.

The second most notable design innovation for the KV7 is in the orientation of the EIDE connectors.

EIDE 1 and 2 lay flat against the motherboard instead of perpendicular. The only reason for this design change seems to be for cable management. Laying the IDE headers flat against the motherboard and at the edge does help stop the ribbon cables from looping out into the case. We feel that the engineers had to place the headers end to end at the side of the board to solve space issues then laid them flat to provide a marketable reason.

The CMOS, USB and SATA RAID headers are all crowded into the lower right corner of the board.

The most downright “goofy” design of the KV7 is the placement of the floppy header. It’s down below the PCI slots along with the two auxiliary fan headers. One fan header should have been on placed behind the backplane for the rear case cooling fan.

The standard five PCI slots which usually go unused in today’s feature rich boards.

CD and AUX in placement hasn’t changed much from motherboard to motherboard. The green jumpers control front audio connections.

The backplane has 10/100 onboard NIC, four USB headers, PS/2 mouse and keyboard, one printer and com port, separate MIC, LINE IN and 6-channel sound connections and a S/PDIF connection. Rarely are com ports used today and cutting one out may have saved a few more pennies.

ABIT does come up with a “who’d a thunk it” design innovation every once in a while. The S/PDIF port has spring loaded doors to protect the connection. (A jeweler’s screwdriver is pushing the doors open in the following image.) This is quite a useful feature compared to inevitable loss of the plastic plugs that other motherboard manufacturer’s use.

The third design flaw is the placement of the power connectors. The main power connector is tucked under the socket area very close to the heatsink. There is a “six of one…half dozen of the other” reasoning why we dislike the placement. The power trunk comes from the PSU and usually loops towards the front of the case then back to the board. The best placement of the power connector is on the DIMM side of the socket and ABIT’s placement on the opposite side of the northbridge chip results in awkward cable management.

However ABIT could argue that the power trunk could be fed across the top of the board then down to the connection thus problem solved….except for enthusiasts with larger heatsinks. This will be explained further on in the review.

The four pin power connection is tucked tightly between capacitors and backplane but how often is a user connecting and disconnection power. It is also highly recommended that a power supply be used that has this power connection cable. Many users in ABIT Forums have recommended this additional power connector be used to provide for more operational stability.

The ABIT KV7 must have presented a few design challenges to the engineers. It is a narrower width ATX motherboard and it is quite obvious that the reduction in space affected the overall layout.

Installation

The ABIT KV7 installs like any other motherboard but three less standoffs are required due to the narrower width. Overall the EIDE headers with their flat-to-the-motherboard design did not present an issue. The design feature did help with cable management by laying the cables flat against the back side of the PC case. Note the way the floppy cable crosses over the PCI slots.

Now to the KT600 chipset itself.

The KT600 has not made a good first impression with us. Another manufacturer’s KT600 platform just plain refused to work. The ABIT KV7 was troublesome right out of the gate until the memory was placed in DIMM 2 and 3. When memory is used in pairs installed into DIMM 1 and 2 the board was prone to desktop crashes. We do not fault ABIT for this as they are working with a VIA product in the teething stage.

The good news is that larger heatsinks can be mounted on the KV7 as evident by the humongous Zalman heatsink. The downside is the location of the power connector. The first problem is cable management as the power trunk line has to awkwardly snake its way around and the location of the power header made the use of the Zalman heatsink almost impossible.

The Zalman heatsink covered a portion of the power header and that meant connecting power prior to installing the heatsink. Realize the Zalman heatsink is an obscenely large heatsink but it is food for thought when it comes to the larger enthusiast heatsinks.

BIOS

ABIT excels with their SOFTMENU bios and it is an overclocker’s BIOS. ABIT seems to be letting enthusiasts have greater control over the BIOS with each new release.

CPU FSB can be set from 100 to 250 with a ratios of 4:2:1, 4:2:1 and 6:2:1. Maximum CPU, AGP, DDR and (new to me) northbridge voltages are shown in the following image.

An interesting feature in the ADVANCED BIOS FEATURES is the DISABLE UNUSE DIMM/PCI CLK setting. Setting this to YES will force the BIOS to detect which DIMM and PCI slots do not have hardware installed and stop sending clock signals to them as if they were not even there.

ADVANCED CHIPSET FEATURES allows for the usual control over RAM and AGP settings.

What caught us in PC HEALTH STATUS was the min/max CPU shutdown temperature. We really liked the broad range.

Also new is improved control over fan failure.

CPU FanEQ will govern fan speed in relation to the ACTIVE TEMPERATURE. If the ACTIVE TEMPERATURE is not exceeded then the voltage is limited to the CPU fan at the percentage set.

Benchmarks…Can KT600 compete?

The ABIT KV7 test system.

• AMD 3200+ 400 FSB
  Processor
• ABIT
  KV7 KT600 motherboard
• ATI 9700 PRO Video Card
  Catalyst 3.4 drivers (Default settings w/VSYNC disabled)
• 2 x 256 MB Corsair PC3200 DDR RAM
• Sony 52x CD
• 60 GB Maxtor ATA133 Hard Drive
• Samsung 950p 19″ Monitors
• USB Keyboard and Logitech USB wireless Optical Mouse
• Zalman CNPS7000A heatsink
• AMK SX1000
  modded PC case (window, fans, cables, loom)
• Enermax 465 Watt FC PSU
• Windows XP Professional Service Pack 1

The Gigabyte 7NNXP test system.

• AMD 3200+ 400 FSB
  Processor
• Gigabyte
  7NNXP motherboard
• ATI 9700 PRO Video Card
  Catalyst 3.2 drivers (Default settings w/VSYNC disabled)
• 2 x 256 MB Corsair PC3200 DDR RAM
• Sony 52x CD
• 60 GB Maxtor ATA133 Hard Drive
• Samsung 950p 19″ Monitors
• USB Keyboard and Logitech USB wireless Optical Mouse
• Globalwin CAK4-76T HSF
• AMK SX1000
  modded PC case (window, fans, cables, loom)
• Enermax 465 Watt FC PSU
• Windows XP Professional Service Pack 1

The ASUS A7N8X v2 Deluxe test system.

• AMD 3200+ 400 FSB
  Processor
• ASUS
  A7N8X motherboard
• ATI 9700 PRO Video Card
  Catalyst 3.2 drivers (Default settings w/VSYNC disabled)
• 2 x 256 MB Corsair PC3200 DDR RAM
• Sony 52x CD
• 60 GB Maxtor ATA133 Hard Drive
• Samsung 950p 19″ Monitors
• USB Keyboard and Logitech USB wireless Optical Mouse
• Globalwin CAK4-76T HSF
• AMK SX1000
  modded PC case (window, fans, cables, loom)
• Enermax 465 Watt FC PSU
• Windows XP Professional Service Pack 1

Programs used

• 
  Sisoft Sandra 2002
• MadOnion
  3DMark 2001 SE
• Quake
  III Arena
• Commanche
  4
• GL
  Excess
• SpecviewPerf 7.0
• Serious Sam SE
• Unreal Tournament 2003 flyby benchmark
• Adobe After Effects 5.5
• SoftimageXSI 2.0.1
• KirbiBench

All tests were run at default video card settings with VSYNC disabled. Bios
was optimized for system but not tweaked for any performance settings requiring
specialized knowledge of overclocking. In other words the BIOS settings were
standard as far as anyone can set the time and date, ensure the ram is 2 or
4 way interleave and at CAS 2 and AGP is chosen as the first boot video card
(8x). The BIOS settings were kept as close to conservative or default value
or as otherwise specified. System settings were set to OPTIMAL (GA-7NNXP) and
AGGRESSIVE (ASUS A7N8X). Memory was kept at DDR400 settings for synchronous
timing with the 3200+ 400 FSB processor. Individual performance will vary with
any particular or specific timings or tweaks enabled by you.

1024 MB page file moved to D: partition. Temporary Internet files moved to
K: partition at end of drive. OS installed to C: and programs installed to
E:. All programs were benchmarked at 1024×768@75Hz with the exception of SoftimageXSI
which requires 1280×1024 resolution.

All motherboards were test “right out of the box”. No special
BIOS timings were set except for the time and date, boot order and health monitor
features or as disclosed.

For all tests an AMD 3200+ 400 FSB processor and two sticks of Corsair DDR400
memory were used.

These may result in lesser or greater scores dependent on variations in system
components and settings. Avoid smoking and excessive amounts of alcohol. Wait
at least 12 hours after scuba diving before flying or driving over 1000 feet
above sea level. Void where prohibited by law. No animals, especially my cat,
were harmed in the making of these benchmarks.

3D Mark 2001 SE

Codecreatures

Splitting hairs; this could be called dead heat.

Commanche 4

GL Excess

Quake III high quality

The Gigabyte 7NNXP is only a stone’s throw from 300 FPS at 640×480.
It’s the closest yet.

Serious Sam

UT2003 Flyby

Sisoft Sandra CPU Arithmetic

It’s a pretty close race with the 7NNXP edging out the A7N8X.

Sisoft Sandra CPU Multimedia

Sisoft Sandra Memory Benchmark

Specviewperf 7.0

SpecviewPerf still grounds itself in the manipulation of 3D graphics
on a business application level rather than on a gaming performance level. The KV7 failed at the PROE-01 test but excelled at DX-07.

The following two tests are targeted mainly towards CPU performance and will
show if any “flaws” are in board design affecting the ability of the
CPU to crunch through the data. While in render mode the two test programs virtually
bypass ram and GPU.

KribiBench

The Kribi engine is 100% software rendering (a pure CPU benchmark) and makes
heavy use of SSE instructions and SMP. More is better.

Adobe After Effects 5.5

Adobe After Effects is a tool to produce motion
graphics and visual effects for film, video, multimedia and the web. It is primarily
a 2D application using imported graphics or digital footage or self generated
effects. A project was created that was a combination of many video footage
files, resizing and rasterizing effects, text animations and multiple layer
effects. This “average” combination was felt to best demonstrate advantages
and/or disadvantages that a real world user may experience rather than isolating
and benchmarking a particular effect.

There is no official benchmark for After Effects
but tasks can be timed to show specific results. Rendering, or the task of building
and compiling frames, is mainly CPU intensive and After Effects generally bypasses
the video card and relies solely on the processor for speed. The time taken
to render 900 frames basically shows how fast the processor is working on the
given task.

The ABIT KV7 trails by 14 and 7 seconds respectively.

Softimage XSI can simply bring
any computer to its knees. It’s an incredibly powerful 3D animation program
that has the ability to become so complex that single processor systems have
been known to “think” for days when rendering an animation. Softimage works
on somewhat similar principle to After Effects. A faster and more powerful video
card will translate to a smoother interface where complex scenes can be manipulated
in real time. Note that Softimage does not have an interface to real-time preview
a finished frame as unlike After Effects. Users can manipulate objects in a
choice of views from wire frame mode to simulated real-time shading mode. In
order to look at a finished frame a user must render the frame to disk which
bypasses the GPU. A faster processor will result in the faster render. The amount
of RAM is not as great an issue as the user is working frame by frame and the
graphics card is doing the bulk of the work while working within the GUI.

This is a most basic overview and there are specialty
hardware components that can enhance the speed and interactivity of complex
3D scenes and programs. The designers working on the test system use Softimage
on a less complex level to provide enhancements and elements to commercials,
promos and station ID elements. Though their work is quite complex to some it
a far cry from that of special effects in major film productions.

Softimage performs its best on a dual processor
system and by far the recommendation for heavy 3D rendering is a dual processor
AMD system.

Unfortunately the KV7 would not run Softimage 2.0.1 and therefore failed the test.

Benchmark conclusions

The ABIT KV7 did manage to pull ahead of competing NFORCE2 boards in some of the tests but the overall results showed some performance inconsistencies in the KT600 chipset. Overall the KV7 does trail but not by a tremendous amount. The most important result is what the benchmarks showed what could not be run; mainly one of the tests in Specview Perf and Softimage. Being a close competitor in some tests is outweighed by not being able to run the race at all in other tests.

Overclocking

Right away I will say that I am not an overclocking fiend. There are many more enthusiasts that are far better at the wonderful world of overclocking than I. In order to give an initial impression I try to max out what I can do within a few minutes of pushing the product. How easy that is to achieve and with what flexibility will determine what potential true masters will have.

The ABIT KV7 has an overclocker’s BIOS. There is no doubt about it due to the many settings available but the KT600 chipset still has some growing up to do and while 12x was stable…12.5 x would not post even at a CPU voltage of 1.8 volts. The before and after:

Quite simply the KV7 has the promise for overclockers to dial in very precise settings but was reluctant to give up its secrets.

Conclusions

ABIT continues to push the design envelope. The KV7 is only really flawed by the power and floppy connector placement and ABIT engineers have done a commendable job packing a lot of features onto an ATX motherboard that is 2.5 inches narrower. The backplane design is efficient. The spring loaded S/PDIF cover is a good touch along with front, rear and LFE connectors that aren’t shared with MIC or LINE-IN. USB ports abound on the KV7 and SATA RAID is forward thinking.

However…

Stability is an issue. Be aware of DIMM usage as we found the KV7 was stable with memory in DIMM 2 and 3 but not in DIMM 1 and 2. The benchmark tests showed a failure to complete one test in Specview Perf and Softimage failed to complete a 100 frame render. This could be attributable to a “disagreement” between ATI video cards and VIA chipsets as many users prefer NVIDIA product for professional 3D modeling applications. This was the first time Softimage failed to function on a motherboard review sample though.

It is our view that the KT600 chipset is still undergoing growing pains. Another manufacturer’s KT600 offering we looked at failed to post altogether. ABIT has always been a favorite of ours and they should not be faulted for VIA chipset teething problems.

The KV7 may turn out a hefty performance return considering the low purchase price. The BIOS is extensive and ABIT documentation is equally as extensive. Is it a “David” to the Goliath NFORCE2? Right now David has only tossed a couple of rocks and they fall a bit short…but not much. KT600 is a better performer than its predecessor but not quite good enough right out of the box to catch NFORCE2. The ABIT NF7-S is the leader in ABIT AMD performance but it also is 20 dollars more at best. 20 dollars doesn’t seem like much in the grand scheme of things but any motherboard manufacturer will tell you it DOES make a difference.

The major plus to the KV7 is in its size. This board could be ideal for smaller PC cases and the KV7 is one to keep an eye on providing specific stability issues can be worked out.

Ratings breakdown.

There are several ratings categories that are used and all may not be applicable
to every product. Our scoring system may not also be “in-tune” with
what you see on other sites. We made it harder to win awards as every product
cannot win the race even if it places well. Just because a product on our site
hasn’t an award does not mean that it is a poor product. Products should fulfill
your needs, have acceptable longevity, and as a pleasant side effect…exceed
your expectations. What you need it for and how it fits those needs changes
with each individual.

Scores Breakdown

Attribute	Score	Comments
Bonus items & software	6	Bare bones.
Design & layout	7.5	The ABIT engineers have done a very good job considering the reduced width but the awkward placement of the floppy and power connectors take away from the overall score. The CPU and Aux fan headers could have been better placed as well.
Documentation	9	ABIT includes excellent hard copy documentation.
Features & options	8.5	Lan, SATA RAID, 6-channel sound, S/PDIF, 4 stock USB + 2 via PCI.
Fine-tuning features	9.5	ABIT SoftMenu BIOS always wins.
Overclocking features	9.5	ABIT SoftMenu BIOS always wins.
Performance & stability	7.5	At default it trails NFORCE2 but not by much. The promise of overclocking “tunability” pulls the score up. To the novice this board is cranky. The instability of professional 3D modeling programs leaves us with a bit of doubt.
Presentation	6	bare bones.
Price / value	8	Quite a lot for a low price.
Total score	71.5/90	79.4%