EPoX 8K9A2+ KT400 Review!!

Mem

Lifer
Apr 23, 2000
21,476
13
81
Over@VR-Zone ,

EPoX 8K9A2+ has a great BIOS with many tweakable settings. The BIOS allows FSB adjustment from 100-255Mhz in 1Mhz intervals, VCore up to 2.0V, VMem up to 3.2V and various memory timings settings. Not to mention, this board is unlocked internally for T'Bred processors therefore multipliers can be adjusted freely.

This board proves to be an excellent board for overclocking and certainly the best choice for overclockers as good stability is still achievable under high FSB overclocking. I am able to increase the FSB from 133Mhz all the way up to 226Mhz and the memory performance is just amazing at that insane speed

 

NicColt

Diamond Member
Jul 23, 2000
4,362
0
71
>Not to mention, this board is unlocked internally for T'Bred processors therefore multipliers can be adjusted freely.

What the F
 

NicColt

Diamond Member
Jul 23, 2000
4,362
0
71
>the EPoX's BIOS engineer told me that there is no need to unlock the Thoroughbred CPUs for this board since the Vss is connected to ground when the T'Bred is seated into the CPU socket

What the F 2

another review at OCWorkbench.com
 

Mem

Lifer
Apr 23, 2000
21,476
13
81
I like the new magic light in the 2nd pic,

The new Magic Light function is first implemented by EPoX into their 8K9A series of boards. The cool blue LED fan sure look really nice when the lights are off and certainly will blend well into a modded case with blue cathode light.




Good for people that have windows and I`m one of those that`s more into modding then OC .


 

Bagheera786

Member
Jun 28, 2000
41
0
0
Did I miss it, or did "OC"Workbench not include overclocking capabilities in their review?

I really hope the folks at VRZone are right about the overclockability...
btw, should I get PC3200 RAM or PC2700 RAM for overclocking?
 

MrEMan

Junior Member
Sep 24, 2002
13
0
0
There are two reasons I haven't chosen either the 8K5A2 nor the new 8K9A, those being that Via is rumored to be readying a VT400A, and also because of EPoX's decision to back off on reading the on-die thermosensor in the AthlonXPs and newer Durons due to confused users returning perfectly good boards because of the more accurate, and thus higher temps they were seeing. I strongly suggest everyone contact EPoX and tell them you want to see the actual temps, instead of totally inaccurate lower temps.

EPoX Tech Support reply concerning on-die temperature readings:

"We do not use the internal thermal diode of the CPU. Too many customers were confused by its higher temperature readings.
Thermal die readings if enabled in the board design would only be accessible to the BIOS (not seen by user) for safey features. I cannot
confirm at this time the thermal die reading ability of the 8K9A board."

Has anyone heard any differently?

Here is an excerpt of an interview done by Van's Hardware with the makers of Artic Silver:


VHJ: How do you explain the fact that many websites do not show nearly this much of a temperature drop when different types of thermal paste are tested? Oftentimes thermal paste roundups show differences of only 1-2 degrees between all the pastes themselves and this variation well within a margin of error.

Nevin: You raise an excellent, excellent point and one I wanted to address. Let?s look at a bit of history:

When Arctic Silver first arrived, it was tested by a variety of websites and yielded results showing temperature drops of 2-7 degrees C -- exactly what we predict. The testbeds used by these websites, however, were primarily Pentium III?s with internal temperature monitoring thermal diodes.

Today, most websites have switched to Athlon-based systems which use a thermistor in the socket for temperature measurement. Unfortunately, this is inherently flawed.

ACCORDING TO BASIC LAWS OF PHYSICS, IT IS IMPOSSIBLE TO OBTAIN AN ACCURATE TEMPERATURE MEASUREMENT DOWNSTREAM FROM THE HEAT SOURCE IN A SECONDARY HEAT PATH IN A NON-ISOTHERMAL ENVIRONMENT (caps added).

Allow me to illustrate further. Imagine a room with one outside window. It?s winter outside?so the temperature outside is 30 degrees, while the room?s temperature is a uniform 70 degrees. A thermistor placed inside the room will, of course, register a temperature of 70?.

Now, imagine what would happen if you took that same thermistor and taped it to the outside of the window. The thermistor will register a MUCH lower temperature than 70?. There is only one section of it even in contact to receive heat flowing from the room, and that heat must pass through the glass. Furthermore, the other 75% of the thermistor is in constant contact with the outside air?which, as we have noted is much colder than the air in the room.

IN SHORT, THE IN-SOCKET THERMISTOR CANNOT EFFECTIVELY MEASURE THE TEMPERATURE OF THE PROCESSOR, ALL IT TRULY DOES IS MEASURE THE TEMPERATURE OF ITSELF (caps added)


Check out the complete Van's Hardware interview with Artic Silver Corp at http://www.vanshardware.com/articles/2001/december/011209_AS_Interview/011209_AS_Interview.htm

 

Tab

Lifer
Sep 15, 2002
12,145
0
76
Originally posted by: MrEMan
Via is rumored to be readying a VT400A

There are already leaked pictures of MSI's KT400A board floating around.

 

First

Lifer
Jun 3, 2002
10,518
271
136
I'd be surprised if we see any KT400A boards in quantity even by December.
 

russr

Senior member
Jun 14, 2000
977
0
0
Perhaps the starter of this thread can also add the OCworkbench review (that was posted in the other thread) to the original post. here's the link: OCWorkbench review

Mr. Eman has a very good point about the kt400a chipset. The truth however is that we do not know when that chipset will be out...it could be January or it could be March or later. The other thing worth noting is that the kt400 vs. kt400a might not be the case of the old kt266 vs. kt266a where there was a siginficant performance difference between the two.

As it stands, right now the Epox 8k9a2+ is possibly one of the best kt400 boards out there and definitely worth considering if you're buying a motherboard right now.
 

Insane3D

Elite Member
May 24, 2000
19,446
0
0
MrEMan

The Epox 8K5A series do use the internal diode of the XP, just not for the temperature monitoring display. The temperature reading in the bios and in Windows is via the in socket thermistor, which as you said is very inaccurate. However, the board does use the internal diode for the overheating protection, and it is set @ 100C. So it does support the diode, just not for temperature monitoring. Epox had a ton of RMA'ed 8K3A series board because people complained they "ran too hot". As much as I wish they hadn't gone back to the in socket thermistor, I can understand their reasoning.
 

NicColt

Diamond Member
Jul 23, 2000
4,362
0
71
>What improvements are made in the 400A?

The only difference will be an improved memory controler that will officially support DDR400 other than that nothing. Even with DDR400 support your still going to have the latency 'issue' because of the bandwidth limiting cpu. Meaning that running in sync 166:166 will probably be better than 166:200 which will just net you about a %5 increase. You'll get %5 increase in speed but also at the price of a %30 increase in latency. In other words the numbers will be better but at the cost of increase latency, it just won't mean much.

The advantage of the NF2 is that you can lock the AGP at 66Mhz and run either ASYNC or SYNC. Also the NF2 and the 8K9A2+ will be able to unlock the processor internally.

Either the NF2 or the KT400 at this time is looking good but with an XP2800. The only reason you would want a KT400A would be if you want to run memory at :200 and accept the latency. If you want to run at sync 166:166 go for NF2 or KT400.

BTW NF2 already has a bug in the IDE drivers and needs an updated driver go figure.
 

KF

Golden Member
Dec 3, 1999
1,371
0
0
>BTW NF2 already has a bug in the IDE drivers and needs an updated driver go figure.
The mobos are out? Somebody has bought one?

The reviewers are saying that VIA needs better memory performance to stay with nForce2, which is why they mention KT400A. If not, they say VIA is in deep **** when nForce2 gets straightened out. Supposedly nForce1 already has 35% of the socket A mobo market.
 

KF

Golden Member
Dec 3, 1999
1,371
0
0
>IN SHORT, THE IN-SOCKET THERMISTOR CANNOT EFFECTIVELY MEASURE THE TEMPERATURE OF THE PROCESSOR,
>ALL IT TRULY DOES IS MEASURE THE TEMPERATURE OF ITSELF (caps added)

OK the socket thermistor is not great, but this sounds like major BS.

The heat divides into different paths according to the thermal resistances. The heat flow in each path is proportional to the temperature. The temperature at the in-socket thermistor is proportional to the temperature at the on-die diode. That is all you need to measure the temperature. You do have to calibrate the system.

When you measure voltage with a voltmeter, the actual voltage being measured is a scaled down voltage which is proportional to source voltage. So this temperature measuring method is actually quite similar to a method that is considered accurate for voltage. Does a voltmeter only measure the voltage of itself?

But the in-socket themistor is better than described anyway. It is close to the temperature of the bottom of the CPU. The poor air circulation within the socket will keep it close.

We can find fault with the on-die diode. Is there some reason the temperatue of the chip would be the same across it's entire surface? How much does it vary? Is the diode next to a hot spot? Next to a cool spot? How do we calibrate the circuit that uses the diode to measure the temperature?

No, the socket themistor is not a terrible accurate in absolute terms, but for measuring temperature differences, such as with AS compared to something else, it should be close. The more likely reason for the drop in the apparent performance of AS in recent times is that the the CPU/heatsink interface has gotten much better due flatter surfaces and higher pressure.


 

MrEMan

Junior Member
Sep 24, 2002
13
0
0
Insane3D,

The FAQ for the 8K5A2 on their site states the following:

"AMD processor temperature detection. Regardless of AMD processor type, the processor temperature is detected via the onboard motherboard thermal sensor."

Is this incorrect/outdated, or does it just apply to the temperature displayed by the BIOS? There is no mention of the on-die thermal diode being used for emergency shutdown only.

Speaking of their site, I wish EPoX would keep it more up-to-date. I notice that NewEgg is selling a EP-8K9A2 board, but EPoX USA's site only lists 4 KT-400 boards: 8K9A, 8K9AI, 8K9A2+ and 8K9A3+.

Also, with all the variations of boards they offer, a comparison chart, including on-die diode support and dimentions around the CPU socket would sure come in handy.


KF,

I am not a thermal engineer, although I used to test heat rises on power line filters. From those experiences I learned that thermal sensor placement is critical. I am curious if you read the entire Van's Hardware interview/explanation with the Artic Silver engineer.


Lastly, anyone had any experience with AOpen boards? Their KT-400 board seems to have a good feature set, including a northbridge fan, and a hardware CPU meltdown protection circuitry. Check them out at http://www.aopen.com/products/mb/ak77-8xn.htm. Their boards seem to offer more features for the same price as EPoX. For example they include 5.1 AC97 audio and LAN, along with the before mentioned northbridge fan for the same price as the EPoX EP-8K5A2. Howvever they have only a one year warranty, unlike the post 9/1/02 EPoX boards which have a 2 year warranty.

One thing that struck me as odd about AOpen, was that in the past they chose to use the VIA chipsets, instead of the ALi chipsets, even though AOpen and ALi are both Acer divisions.

Thanks for the replies.
 

NicColt

Diamond Member
Jul 23, 2000
4,362
0
71
>The reviewers are saying that VIA needs better memory performance to stay with nForce2 If not, they say VIA is in deep **** when nForce2 gets out.

Well not really. Let's compare oranges to oranges. VIA does not 'need' better memory performance, it will do one but does not need it. Yes the NF2 board does look good and will have a nice feature set but to say that NF2 will seriously outperform VIA by it's memory performance is a little strong. Anandtech has shown that in real world performance is so small that it's not even worth talking about it. Ok it does have official DDR400 support but again It's only about a %5 increase. Once the KT400A comes out with an updated memory controller it will most likely just match it. Then again SiS 746FX will most likely own all, mark my words.

Like I said before, even with an official DDR400 board you now have to deal with the latency issue. The better the memory the greater the latency problem will be.

VIA's big mistake was to name their chipset KT400 when they should have named it KT333A. KT400 to most has the appearance of supporting DDR400 which officially they don't. Since JDEC still didn't approve an official DDR400 standard there was and there is no rush to implement an official DDR400 chipset.

Sadly though this is the end of the line for this generation of processors and chipset in this class. With Granite Bay and Hammer coming out an entirely new generation of chipset will take over and we'll be privy to a whole new ball game.
 

KF

Golden Member
Dec 3, 1999
1,371
0
0
AS interview

>I am not a thermal engineer...
Me either. Not remotely. My quarrelling is strictly on the basis of the logic which has been presented.

>...From those experiences I learned that thermal sensor placement is critical.
No doubt you are right. I think what people find out when they measure things is that it much more difficult than imagined to represent anything definate with a single measurement.

> I am curious if you read the entire Van's Hardware interview/explanation with the Artic Silver engineer.
No, but now I have. It is a good, informative article. It is also a sales pitch for AS.

I recall reading elsewhere similar statements over the years quoted from the AS guy (Nevin House). Here he speaks at length for himself. I have also read volumes of this type of discussion over the years. Almost all of it is going around in circles over the same thing.

What Nevin is calling compression, I would call proportionality. Yes the temperature differences are lower, but they are proportional. Anyone who does any measuring realizes that you do not deal with the original source: you transform it to something proportional. Consider even the exemplary on-die diode. It will generate a voltage difference that is proportional to its temperature. Obviously it doesn't go up 50 volts if the temperature goes up 50 degrees C. If you were to add 50 degees to the actual voltage rise (possibly .01 volt) to supposedly get the true temperature, it would be utterly absurd, and utterly wrong.

Nevin says that mobo makers simply add a constant to get the temperature to look like the true (die) temperature. If that is what they do, then the temperature difference is compressed. If on the otherhand they multiply by some factor, as they should, the temperature is equal.

If you look at it carefully, you see that Nevin's argument against the validity of the in-socket thermistor, is actually a demonstation that it an accurate proportional analog.

The in-socket thermistor is also very close to the source. Therefore you do not introduce large inaccuracies due to a tiny difference multiplied by huge factor, as you would by measuring the temperature of window pane on the outside in order to derive the temperature of the inside. Also there is not a large temperature gradient across the thermistor, as there would be in the window pane example.

All this pertains to the claims for AS. The difficulty of determining the real (die) temperature when different heatsinks are attached remains a problem because you change the way heat divides into different pathways.

 

MrEMan

Junior Member
Sep 24, 2002
13
0
0
KF,

Thanks for your comments regarding CPU temperatures.

I still would encourage EPoX and other board manufacturers to display both temps on the BIOS screens, and also to provide a utility to display both temperatures real-time in Windows and LINUX. Only then would we be able to perform an accurate analysis of the relationship between the two measurements and determine whether or not a linear or non-linear multiplication factor can be determined to accurately model the differences between the two (for knowledge, and for use by those with boards which only measure the CPU temps from the under-socket sensor). The accuracy and tolerances of both sensors would also need to be known for the analysis to be as accurate as possible.


Does anyone know how the temperatures are read and displayed on Intel P3/P4 systems?


Lastly, twice in the last 2 days, my reply has totally vanished as I was in the process of typing it. Is anyone else having problem with the Anandtech forums?

Thanks

MrEMan ( -> Mister E Man -> Mystery Man)
 

KF

Golden Member
Dec 3, 1999
1,371
0
0
>Lastly, twice in the last 2 days, my reply has totally vanished as I was in the process of typing it.
> Is anyone else having problem with the Anandtech forums?
It hasn't been happening to me. I sometimes take an hour or so to do some. There is a Forums Issues section that
might have info on this.

It sometimes will take 15 minutes to do the sending, or at least acknowledge it. I can see nothing is happening because I watch the lights on the external modem. But even when I have been disconnected, the message has showed up.

 

DrPara

Member
Jul 8, 2001
68
0
0
i think the only good thing the kt400 is bringing to the market is 1/6 DIVIDER
2 bad 8k5a2(and 8k5a3) series dont have a 1/6 PCI DIVIDER
 

Live

Member
Oct 20, 2002
90
0
0
Originally posted by: DrPara
i think the only good thing the kt400 is bringing to the market is 1/6 DIVIDER
2 bad 8k5a2(and 8k5a3) series dont have a 1/6 PCI DIVIDER

There is no support for 1/6 divider in kt400. Some mobo makers stated it had a 1/6 divider (ASUS and Soyo I believe) but that was just crap. No board with current VIA chipset has it nor will they ever since its hardware related and not software. We will just have to wait and see if a kt400a or the like will have it. It?s stated that nforce 2 has it but they said the same thing before with kt400. Hopefully they are right this time.
 

DoubleL

Golden Member
Apr 3, 2001
1,202
0
0
Well I don't have anything better to do right now so I ordered a 8K9A2+ to see what it will do with a XP1900, 512 mags Corsair 3500 and a couple WD80,s
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |