 |  | | 
|
New Abit KA7 beta-BIOS
|
9th May 2000 |
|
Posted by: Andy
I've been sent the very latest Abit KA7 beta-BIOS which gives L2 cache latancy adjustment within the BIOS! As always I do not recommend you use beta-BIOS and you are on your own if it goes wrong. However, for those who want to have a play, here it is.
These are the new features.....
The Latest KA7 BETA BIOS
Add Level 2 Cache ID Selection in SoftMenu III
With this Selection , you can determine Level 2 Cache divider
There are eight items
Default , 3 , 4 , 5 , 6 , 7 , 9
Some important technical info here... if your cache uses SRAM made by SEC or NEC, choose these options....
If want to make Level 2 Cache running at 1/3 frequency of CPU frequency
, Set Level 2 Cache ID to 3. If want to make Level 2 Cache running at 1/2 frequency of CPU frequency
, Set Level 2 Cache ID to 5. If want to make Level 2 Cache running at 2/5 frequency of CPU frequency
, Set Level 2 Cache ID to 9
If if your cache uses SRAM made by Mitsubishi or Motorola
If want to make Level 2 Cache running at 1/3 frequency of CPU frequency
, Set Level 2 Cache ID to 4.If want to make Level 2 Cache running at 1/2 frequency of CPU frequency
, Set Level 2 Cache ID to 6. If want to make Level 2 Cache running at 2/5 frequency of CPU frequency
, Set Level 2 Cache ID to 7
Currently, you need to clear the CMOS if using the wrong divider causes boot
problems... Abit are aware of this problem and are working on a fix.
 Daily Hardware News
|
|
Gigabyte GA-7VX KX133 Motherboard Review
|
3rd May 2000 |
|
Posted by: Andy
Following on from last weeks KX133 Motherboard RoundUp
I've posted a standalone review of the Gigabyte GA-7VX.
I should be updating the round-up over the coming weeks, Soltek for one want their board included and are sending a sample of their newest revision. I've also been sent an early version of the Soyo SY-K7VIA and while its not fit for review yet IMHO, its looking a nice board, and could well steal a march on the Abit KA7 as king overclocker :)
Gigabyte GA-7VX
Daily Hardware News
|
|
KX133 Motherboard RoundUp
|
27th April 2000 |
|
Posted by: Andy
I've posted a comparison of no less than six AMD Athlon KX133 motherboards, including the the Tyan Trinity K7, Asus K7V, Gigabyte GA-7VX, AOpen AK72, and Epox 7KXA!
Overclocking performance, application speed, memory and gaming scores are included, in an attempt to find winners among the boards. For those of you considering a KX133 you may be surprised by my closing comments in the article....
KX133 Motherboard RoundUp
Daily Hardware News
|
|
Tyan Trinity K7 KX133 Motherboard Review
|
25th April 2000 |
|
Posted by: Andy
I've upload a (fairly short) review of the Tyan Trinity K7 KX133 Athlon board. This offers good expansion, and on paper, support for overclockers, including bus speed selection and voltage tweaks. How does it fair in practice.....
I'm hoping to get the K7V review up soon, as I know many of you want to see how this compares with the Abit KA7, but I've lost half my test results, and 2/3's of the review text in a slight accident (cough!) with a Windows 2000 installation, so I need to re-do some of the tests :(
Tyan Trinity K7 KX133 Motherboard Review
Daily Hardware News
|
|
MSI 6301 i820 (Camino) Motherboard Review
|
22th April 2000 |
|
Posted by: Andy
The review of the MSI 6301 i820 (Camino) motherboard review is now online. This board from MSI uses DIMM's rather than RAMBUS, so it makes this new Intel chipset, with AGP 4x, and UltraDMA/66 available without re-mortgaging your house due to the high cost of RAMBUS.
MSI 6301 i820 (Camino) Motherboard Review
Daily Hardware News
|
|
Chipset Benchmark Comparison
|
20th April 2000 |
|
Posted by: Andy
Taking a break from motherboard reviews, I thought it would be helpful to
establish a benchmark for performance from the new chipsets on the block,
particular i820 (Camino) and VIA Apollo Pro 133A.
I've run an extensive range of benchmarks across 4 mainboards, each
sporting a different chipset. I've looked at memory, AGP, gaming and
application performance. The i820 board appearing in this review, will be reviewed here within 48 hours, with the Asus P3V4X coming shortly after.
Intel i280 / VIA Apollo Pro 133 / VIA Apollo Pro 133A / Intel BX Chipset Benchmarks
Daily Hardware News
|
|
Upcoming Stuff....
|
18th April 2000 |
|
Posted by: Andy
Thanks for all the kind comments on the Abit KA7 review! Its good to know I've still got it ;) A few of you have asked why I picked up things that were not mentioned in other reviews. I can't really comment as I never read other reviews, and just write what I find. I do think the fact that we don't churn out a review a day, does mean we can be more thorough than some other sites. There's a hell of a lot more to doing mobo reviews than running Quake3 and writing "Wow!" :)
Right now I'm working on a number of reviews. In no particular order Tyan Trinity K7, MSI 6301, Asus P3V4x, Asus K7V, various Soyo boards, and a couple from Gigabyte. Looks like I'll be busy! I'm also working on a couple of non-review articles, something I've not done for a while, but should be interesting! I'm hoping to have at least one new review and a new article up by tomorrow.
Daily Hardware News
|
|
Abit KA7 Review
|
13th April 2000 |
|
Posted by: Andy
After 4 days and 3 nights of continuous testing, the Pro Plus and Coffee is wearing off, and this tired hardware tester needs sleep!
However I put together the Abit KA7 board review, and although its a nice board, we've found some issues during testing that really need your attention. Want to run your Athlon at 133Mhz or over? Forget it - so is this the board to use when 133Mhz Athlons become a reality? We also look at some memory issues as well.
What the verdict? Well despite the issues, we really like this board, which was made for the overclocker! If you have any inclination of overclocking your Athlon you need to look at this board.
Abit KA7 Review
And now I will sleep.....
Daily Hardware News
|
|
Nearly there with KA7....
|
13th April 2000 |
|
Posted by: Andy
I've put in 3 all-nighters on the KA7 testing, so I am going to see if I can get the review written today :) I've been playing with a beta-BIOS as well - this board definately has issues with SDRAM which I am hoping this will fix. I also found something today about KX133 that arose from KA7 testing that nearly blew me away - VIA you have done a very bad thing.....
OK well, provided I can stay awake, I'm going to try to get the review up by midnight tonight GMT. I'm still testing right now, working through the issues, so no words written at all yet - if I can string together something semi-articulate after 3 days of all nighters (does that make sense...?) there should be something up here today. Be warned, I do test VERY thoroughly and you will see issues and problems that won't get mentioned in any other reviews.....
Daily Hardware News
|
|
More Abit KA7 Testing
|
12th April 2000 |
|
Posted by: Andy
Testing on the Abit KA7 has been continuing, and I'm feeling rather sleepy after an 18 hour stint on this. Results are looking very nice now, and I've found a sweet component combination that is even allowing me to boot the 0.18u Athlon 650 @ 815/125! Not bad at all, and although things are not 100% stable at that speed (almost certainly due to the L2 limitations of my test CPU), its seems to be 100% stable at 6.5 x 120, which is the best I've done on any board to date!
I do have concerns about memory scores, these are much lower than the ones I've obtained on my reference VIA Apollo Pro 133A system, although using the +33 PCICLK option, gives a decent enough boost. But overall the board is looking very nice, and I think it is going to displace the Epox EP-7KXA in my own personal system. One thing is for sure, without wanting to pre-empt the review, this is the board that Athlon overclockers should choose.
While I am talking about VIA mainboards, currently they are all using the 686A Southbridge with UltraDMA/66. It may be worth holding off on any purchase for now (even the KA7) as the 686B Southbridge is ready any day now, and is drop-in pin compatible with the 686A meaning many boards will be coming with UltraDMA/100 capabilities for free very soon! Worth a thought....
Daily Hardware News
|
|
Abit KA7 Provisional Findings
|
11th April 2000 |
|
Posted by: Andy
I've been playing around with the Abit KA7 for around 12 hours, and this is no way enough testing to even think about writing a review. I am getting a lot of mails asking what I think, in the light of my upcoming Asus K7V review so I'll pass on a provisional report...
The board layout is nice, and despite the 4 DIMM slots, they are is still plenty of clearance between SlotA and DIMM slot1. Its certainly possible to use even the biggest Athlon coolers. Performance is pretty much the same as other KX133 boards I've tested, as you'd expect memory scores are way down on Intel chipsets, but performance is par for the course for KX133.
As an overclocking board - well so far, the best I've done is 109Mhz with a 0.18u and 0.25u Athlon 650. This is 2 Mhz better than the K7V, but 6 Mhz -worse- than the best I can do with the Epox EP-7KXA , which runs at 745/115 Mhz effortlessly at 1.75v on the 0.25u Athlon, and at 1.65v with the 0.18u Athlon.
The board is also exhiting pickyness with RAM, and won't work with all 4 DIMM slots populated with PC133 memory from different manufacturers. This is not necessarily any worse than any other KX133 board I've tested in this regard.
I hope you will bare with me on this review. I do like to test for at least 7 to 10 days, before I commit anything to paper, as a rushed review is not doing anybody any favours. In the meantime, if you need a KA7 fix, I will post links to other reviews as they arrive on my Daily Hardware News page.
BTW while looking at the KA7 box, I saw BXBoards logo on the side, and frankly it looks a load of old rubbish compared to all the other ones! (SysOpt in particularly have a nice one!). So if anyone fancies doing a nice logo for the site to replace the one at the top of this page, send it to bxboardshaveacraplogo@bxboards.com and I will send something nice to the one I chose :) Get those Crayola's out chaps....
While I am doing provisional reports, the Tyan Trinity K7 is also looking nice, not really an overclocker though, as it only has 180, 200, 240 and 266 Mhz bus speeds. I suspect a newer BIOS may reveal some other settings as happened with the Trinity 400 however.
BTW on a different note, many of you are asking what I plan to do about the domain name! Well we've done a hell of a lot more than BX in our time, and BX actually stands for Bad Ass eXtreme, not BX as in BX chipset ;)
Daily Hardware News
|
|
VIA Announce DDR SDRAM Support
|
10th April 2000 |
|
Posted by: Andy
Many of you are curious about the board I mentioned with voltage tweaks, that did not work (see below)! What happened was that a few engineering samples escaped into the wild. They were not fit for sale or review so BXB contacted Soyo, sent a diagnostic report, and sat on our review until the final production run was done. For those of you who have obtained a Soyo SY-7VCA and find voltage tweaks don't work, shout at your retailer!
To check if you have an affected board, check your BIOS so see if a 10% voltage increase shows any effect (ie 10% adjust for a CPU with a default of 2.0v will show up around 2.2v). If selecting voltage adjust does not raise processor voltage, you have an engineering sample! The fix can be downloaded from BXB - 7VCA Vcore fix. This is for informational purposes only, and should not be attempted!! Again, I'd like to reassure you that only 80 of these boards made it to market, and they should not have been sold or reviewed.
On a different note, VIA have finally announced what has been an open secret for some time - support for DDR SDRAM in future chipsets.
Fremont, CA, April 10, 2000 - VIA Technologies, Inc., today announced it has selected Double Data Rate (DDR266) SDRAM as the memory technology that will follow its successful PC133 for future PC chipsets. DDR266 supported VIA chipsets will be targeted across the PC Market at servers, mobile and, desktop systems, s and Value PCs and mobiles. Established on the existing SDRAM industry infrastructure, DDR266 is a rational high performance and cost-efficient solution that allows easy adoption for memory vendors, OEMs and system integrators. In cooperation with key memory suppliers including Micron, Samsung, Hyundai, Infineon, Hitachi, NEC, Mitsubishi, Nanya, Toshiba and others, VIA is proactively participating in every stage of the engineering development of the DDR266 device and PC2100 memory module technology to ensure that VIA's DDR chipset is compliant with current JEDEC DDR SDRAM standards.
"DDR266 is a sensible evolutionary solution for the PC industry that builds on the existing SDRAM infrastructure, yet makes awesome advances in solving the system performance bottleneck by doubling the memory bandwidth," said Wen-chi Chen, President and CEO of VIA. "With the PC133 development effort, VIA established itself as a leader in promoting appropriate memory technologies for the PC industry, and we aim to continue in that role by enabling the PC platform's transition to DDR266."
snip
DDR SDRAM will offer a superior solution and migration path from existing SDRAM designs due to its availability, pricing and overall market support. DDR266 doubles the data rate through reading and writing at both the rising and falling edge of the clock, achieving data bandwidth 2X greater than PC133 when running with the same DRAM clock frequency. With peak bandwidth of 2.1GB per second, DDR266 enables system OEMs to build high performance and low latency DRAM subsystems that are suitable for server, workstation, high-end PC and value desktop SMA systems. With a core voltage of only 2.5 Volts compared to conventional SDRAM’s 3.3 volts, DDR266 is a compelling solution for small form factor desktop and notebook applications.
On paper this looks an interesting, although logical evolution of existing SDRAM technology, and the key note peak scores look higher than RAMBUS, which could threaten the future of this over-priced, and much maligned, alternative RAM technology further out in the cold in the coming months.
Asus K7V review is nearly done, so far its not the overclocking champ I'd hoped, as its only run the Athlon 650 up to 699 (6.5 x 107) and this is with 0.18u and 0.25u flavours of the chip. Despite that, its still a quality product and I hope to get it written up for posting late today UK time.... this review will be the last in the current BXB style, I am currently evaluating adding additional bench tests for future reviews at BXB.
I've also just received the Tyan Trinity K7 and the damn thing has voltage tweaks. I nearly fell off my chair when I saw Tyan had done this....with 6 PCI slots, and some bus speeds too, looks like Tyan could be a dark horse in the Athlon OC stakes...
Daily Hardware News
|
|
A Day in the life of a Hardware Reviewer
|
7th April 2000 |
|
Posted by: Andy
BXBoards get to see a lot of hardware, and sometimes it doesn't quite work! I'm not going to reveal the board or the manufacture here, but I was testing a Socket 370 VIA Apollo Pro 133A board, and noticed that it was not a stable overclocker, even though their previous boards in the past have been awesome!
More investigation revealed that the voltage tweaks in the BIOS did not actually work, so the Celeron 300a @ 504/112 and Celeron 366 @ 605/110 was failing. These babies need 2.1 or 2.2v to be 100% stable, so it was BSOD time during testing at these speeds.
Quick mail to xxxx (whoops, nearly said it), with a fault list, which I what I always do in this situation. After fielding frantic e-mails from them, they sent me another board! Uh.. guys, still broken. Voltage tweaks do not raise the voltage!
More panic follows! Then I get a word document in my e-mail titled "xxx Vcore rework instruction". In a nutshell what I needed to do is to remove a 750k-0603 resistor from the first board, and solder onto the new sample and all would be fixed! Hell.....
If you have worked out what board I'm talking about, award yourself a Gold Star. Rest assured, the board I have is not representive of the final product :)
On a less trivial note, the Asus K7V review is coming this weekend.....
Daily Hardware News
|
|
KX133 Stability Problems?
|
6th April 2000 |
|
Posted by: Andy
I've been getting a lot of mails complaining of instability with some KX133 boards, notably the Epox EP-7KXA VIA KX133, which is my current board I use in the machine I write all the site reviews! So I'd be the first to know if the board had problems, as its in use 24/7 for SETI, DreamWeaver 3, and the odd Quake 3 game :)
Much of the instability people are seeing is a direct result of using unofficial beta nVidia drivers. For Win2K I use the 3.78 RC drivers, with no problems. With anything else I get lockups. For WinNT I use 3.68 Detonator drivers, again with no problems. For Win95/98 take your chances, personally the OS is a joke, with its inherant non-support for any chipset that is not Intels. However again 3.68 Detonator looks a good bet!
Personally I've been running the 7KXA with an Athlon 650 running at 745/115 for over 3 weeks now, with not one lockup or crash under Win2k :)
Daily Hardware News
|
|
AOpen AX64Pro Review
|
4th April 2000 |
|
Posted by: Andy
Having been pretty disappointed with AOpen's recent Athlon mainboard, we
were granted a look at their AX64Pro mainboard, with the VIA Apollo Pro133A
and unlimited voltage tweaks in the BIOS.
On paper it looks like a board to regain AOpen's crown for overclocked
stability and quality, but have they managed it in practice?
BXB took the AOpen AX64Pro round the block and came up with some surprising
results....
AOpen AX64Pro Review
Daily Hardware News
|
All content and graphics on this site are protected by U.K. law and international treaties and may not be copied without the express permission of Andrew Drake, who reserves all rights. Use of any of
BXBoard's content and graphic images are strictly prohibited.
All company trademarks are acknowledged. All rights reserved.
| |
