ECS K7S6A Review, + 1 more from t-break - 74%

[Athlon4all]

<< You lot are a bunch of w*****s [those who dish this board , that is], this board sells at peanuts, offers the stability of the best boards available, and is only a few % (and virtually undetectable when in use) slower than kt266a chipsets...!!! Benchmarks may show nice numbers but lets face it :

1. Who here can notice 10 - 15 more FPS frames in quake 3 when the average framerate is above 85hz?? Honestly if you can... >>

I agree about the benchmarks. nForce, 266A, and 745/735 are pretty much on even grounds, what I don't agree on is the stability/compatibility, but this is purely ECS' doing, 745/735 are great chipsrts, but ECS has done very poor work with them, *sigh*. Waits for Abit KS7 and others.

 

[NFS4]

<<

<< http://www.tbreak.com/hard/mobo/ecs_k7s6a/page1.html

Poor thing got beat up by a "lowly" KT266A in most of the benches...even at 166MHz memory speeds. Pitiful

>>



You lot are a bunch of w*****s [those who dish this board , that is], this board sells at peanuts, offers the stability of the best boards available, and is only a few % (and virtually undetectable when in use) slower than kt266a chipsets...!!! Benchmarks may show nice numbers but lets face it :

1. Who here can notice 10 - 15 more FPS frames in quake 3 when the average framerate is above 85hz?? Honestly if you can...

2. Why pay double the price for a board that's only fractionally better, and is going to ( in the months to come ) go out of date???

So before you go slaggin off chipsets about its performance and so on, OPEN YOUR F****G EYES!

Edit: No im not against Via or Intel or any other brand, i think its silly how you people seem to quickly forget that a few percentage points makes all the difference. Doesnt the WHOLE image count as something? >>


1) The Shuttle AK31 Rev 3.1 (KT266A) costs $7 more than the ECS K7S6A and doesn't need DDR333 to run respectably.
2) It also has 6 PCI slots and 4 DIMM slots
3) It has a whole host of overclocking/voltage/memory tweak options
4) It doesn't require more expensive DDR333 to run fast
5) It has no problem with multiplier locks or FSB speeds
6) The ECS K7S6A gives up the onboard LAN (over the K7S5A) while not giving an additional PCI slot...STUPID!:|
7) As for performance, the fact that this thing hardly performs any better than the K7S5A is the nail in the coffin

As for compatibility and stability, the SiS based K7S5A has been a horror story around these parts.

 

[rockhard]

Heres a link from the Inqwell - hmmm?
The Inquirer - pinch of salt

Cut n pastey from HOCP:-



<< Mike is dead on with this information. Benchmarks we have seen show only a 1% to 3% difference in scoring. The fact of the matter is that the 266MHz bus of the Athlon is getting saturated beyond 266MHz memory bus unless the CPU bus scales with it.

VIA is going to have a tremendous uphill sales job with this chipset. Add to the fact that there still is no official PC2700 JEDEC specifications just makes the whole thing a bit nastier.

Here are some numbers that we are not under NDA with to show you exactly what Mike is talking about. It compares a KT266A board to a equally equipped KT333 board. Can you tell which is which? We are not going to tell you, but I bet you can figure it out.

Spec ViewPerf version 6.12

Awadvs-04 49.77 49.79
DRV-07 20.30 21.11
DX-06 30.58 30.58
Light-04 9.512 9.848
MedMCAD-01 28.95 29.03
ProCDRS-03 14.29 14.29

Feel free to post our proof Mike!
>>



Marketing with numbers anyone?
Intels tactics spreading?

 

[NFS4]

<<

<<
1) The Shuttle AK31 Rev 3.1 (KT266A) costs $7 more than the ECS K7S6A and doesn't need DDR333 to run respectably.
2) It also has 6 PCI slots and 4 DIMM slots
3) It has a whole host of overclocking/voltage/memory tweak options
4) It doesn't require more expensive DDR333 to run fast
5) It has no problem with multiplier locks or FSB speeds
6) The ECS K7S6A gives up the onboard LAN (over the K7S5A) while not giving an additional PCI slot...STUPID!:|
7) As for performance, the fact that this thing hardly performs any better than the K7S5A is the nail in the coffin"

As for compatibility and stability, the SiS based K7S5A has been a horror story around these parts. >>



If i remember so, you had problems. You slagged it down in particular.

HENCE YOUR Of course, that's just my opinion; I could be wrong. HAHAHA , why bother writing anything if you cover you jacka*s with that?

End of debate. >>


What is that? I list 7 reasons why this board is not a viable alternative to current SiS735 boards or KT266A boards and all you an do is cuss and make rude comments? Go play in your sandbox, the children are waiting.

 

[Pabster]

Not to be ridiculous, but let's not exaggerate here. 745 A1 at 133/133 scores within 1% of KT266A at 133/133, and, will likely surpass it easily with a decent BIOS.

Perhaps K7S6A is not a viable upgrade from K7S5A, depending on your situation. If you aren't utilizing the on-board features (and many are not), would like the extra DIMM (and extra capacity), and want a performance increase to boot ... the extra $13 is hardly anything to bitch and moan about. And, I remind you that K7S6A does not require PC2700 modules to reach full potential. Again, take a good look at the RAM scores at 133/133. It's very likely another vendor will significantly improve on ECS' scores, or, perhaps, ECS will further optimize the BIOS as they've done with K7S5A over the past several months.

I don't know about you, but I'll take a 745-based mainboard ANY DAY over KT333.

 

[NFS4]

<< Not to be ridiculous, but let's not exaggerate here. 745 A1 at 133/133 scores within 1% of KT266A at 133/133, and, will likely surpass it easily with a decent BIOS.

Perhaps K7S6A is not a viable upgrade from K7S5A, depending on your situation. If you aren't utilizing the on-board features (and many are not), would like the extra DIMM (and extra capacity), and want a performance increase to boot ... the extra $13 is hardly anything to bitch and moan about. And, I remind you that K7S6A does not require PC2700 modules to reach full potential. Again, take a good look at the RAM scores at 133/133. It's very likely another vendor will significantly improve on ECS' scores, or, perhaps, ECS will further optimize the BIOS as they've done with K7S5A over the past several months.

I don't know about you, but I'll take a 745-based mainboard ANY DAY over KT333. >>


I'll give you this Pabster. It is decent at 133/133. But this board was made for 133/166 operation otherwise they would have just stuck with the SiS735. I am just dissappointed with the 133/166 scores which require the use of more expensive PC2700 memory.

If you are using a ECS K7S5A or a KT266A motherboard, then this board is not anything to jump to. That's my whole point. It just doesn't prove itself. Besides, SiS745 and KT333 are dogs anyway IMHO. The fun doesn't start happening until 746 and KT333A.

Anyway Pabster, how are you enjoying your K7S6A? Was it worth it coming from a K7S5A?

 

[LXi]

tdas2 is the perfect representation of how desperate ECS homage payers are right now. Your dearest hardware had betrayed your expectations, so just flame everyone for rightfully overlooking this lackluster product.

 

[Pabster]

NFS4 wrote:

"If you are using a ECS K7S5A or a KT266A motherboard, then this board is not anything to jump to. That's my whole point. It just doesn't prove itself. Besides, SiS745 and KT333 are dogs anyway IMHO. The fun doesn't start happening until 746 and KT333A."

No argument, but, again, not everyone is going from 5A to 6A. That said, 6A is hardly a bad choice. The small price premium might be worth it (to some) for the third DIMM slot alone. And, 745 does outperform 735. So for $13 more, you get a few more points in benchies and more RAM.

"Anyway Pabster, how are you enjoying your K7S6A? Was it worth it coming from a K7S5A?"

It's not bad at all. I'm running an XP 2000+, all stock. I don't have to choose between the boards, because I didn't get rid of any of my 5A's. Why would I?

 

[gogeeta13]

Brandon, the only people who would utilize the 166, would be enthusiasts. and if you are an enthusiast, you would know, asynchronous mem and fsb is the worst.

If someone on these boards has a change to test it at 166/166/33, then we will see. Other than the Iwill XP333, it will be the only other socket423 board that can run the PCI and AGP in spec.

Let's wait till we get 166/166/33 benched before we jump to conclusions

 

[NFS4]

<< Brandon, the only people who would utilize the 166, would be enthusiasts. and if you are an enthusiast, you would know, asynchronous mem and fsb is the worst.

If someone on these boards has a change to test it at 166/166/33, then we will see. Other than the Iwill XP333, it will be the only other socket423 board that can run the PCI and AGP in spec.

Let's wait till we get 166/166/33 benched before we jump to conclusions >>


If the only people that would utilize 166 are enthusiasts, then what is the point of the SiS745 then?

The only other thing that it adds that the 735 doesn't have is integrated Firewire.

We have to wait until 746 to see AGP8x, ATA-133, and USB 2.0.

 

[Insane3D]

Ummm...maybe I'm missing something here, but the Sis735 can do 133/166/33/66, 166/166/33/66, and even 200/200/33/66. The Leadtek board has all these options in the bios. Of course, it's another story getting these settings to work..

 

[nortexoid]

NFS4, i hope it's no surprise to u that companies often thrive off marketing gimmicks like...

can u guess?

what's the big deal?

 

[NFS4]

<< NFS4, i hope it's no surprise to u that companies often thrive off marketing gimmicks like...

can u guess?

what's the big deal? >>


Well of course I know that, DUH! But for someone here to use that as an excuse to justify the performance is beyond me.

 

[Pabster]

There is no marketing gimmick!

SiS 745 outperforms 735 across the board. Even at 133/133. That's blatently obvious, at least in the synthetic benchmarks so many people here seem to thrive on.

PC2700 is no gimmick either. Running asychronously is never a good idea for maximum system performance. That said, the A1 revision 745 does not initiate additional wait states as 735 did -- which leads me to believe the slightly lower performance in some areas on K7S6A @ 133/166 is a BIOS-related issue. We'll see as the rest of the 745 boards appear.

746 is definitely something to look forward to, but to call 745 a "marketing gimmick" ... that's asinine. Especially at a $70 price point. I think most people would agree that (performance notwithstanding) the extra DIMM and overclocking features are well worth the few bucks over K7S5A. YMMV.