[H]ardOCP ASUS x32 Review

malicious

Supreme [H]ardness
Joined
Apr 8, 2005
Messages
4,772
I don't understand why the gameing benchmarks are the way they are w/ mobo reviews here at the [H]ard.

It's so hard to make a direct comparison between boards.

At least I know that an FX57 in a x32 is much faster than 3500+ in an sli-deluxe.

Why the confounding variables?
 
Well, at least they kind of put a trail you can follow...the A8N32 with 4800+, then the KN1 with 4800+, then the KN1 with 4000+, then the A8N-SLI with 4000+, then the A8N-SLI with 3500+ :). I suppose if you want to, you could come up with a formula based on that info that would give you A8N32 with 3500+ performance.

That being said, I would guess that they use the 2-3 newest processors they have lying around. This case might be a little different because the A8N-SLI was probably reviewed a long time ago. Rather than rerun all the tests using the same old processors, they run on probably the 2-3 newest ones they have and put up a bunch of results. I agree it's kind of hard to compare apples to apples, but it's very likely a time issue.
 
It's not like they are gening out reviews every day.

Speaking of apples to apples they took that section away from GPU reviews.

I love the [H]ard and [H]ang on their every word but I go elsewhere for comparing component performance.

look at their ati review and tell me which card is faster by the charts.
 
You guys make a good point on the CPU comparison.

Really, we are comparing board features, overclockability and stability above all else. The A8N-SLi Deluxe review numbers were added in to compare the A8N32-SLi Deluxe vs. the older model. Really to show what was improved upon.

In the review, I a couple comparisons between the older boards and the newer one. Mainly to outline what is improved on. I think that many of the people looking to buy this board are owners of one of the earlier A8N-SLi boards. So for them what's changed is of paramount concern. Though being much older, you can't easily compare CPU scores to a newer board tested with a newer CPU.
 
Well, my thought's on the whole CPU comparison are that with the on-die memory controller, there isn't more than say 1-3% difference between boards using the same processor.

One of the main differences I see influencing overclocking and stability is the bios revision. A good case being the Abit AN8 Ultra board. It kind of got a negative review at [H], but it's worked fine for most people after say the 18 bios or so.

Is there a reason the OC'd benchmarks were dropped? I remember back a few reviews back where the board/processor then say 250/500 for bus/memory speeds. Just wondering.

I think your reviews do a great job of pointing out the features, including pictures of the accessories and stuff like that as well as the bios. It has been helpful for me anyway :).
 
Not to be rude but certainly here to speak my mind: I have found the [ H ] reviews to become increasingly infrequent, late, and near useless. The lack of competing product comparisons is inexcusable. I find myself visiting here more out of habit then of obtaining any useful information. I hole heartedly agree with the testing methodology for GPU's but that is where I draw the line. When I am shopping for a motherboard I need to see comparisons to like products, the more the better. This board has been reviewed all over the net and I have already formed my opinion a month ago or more that upgrading from my A8N-SLI Deluxe would be near pointless. I also feel that these "PREVIEWS" are poor excuses for not getting a full review out in a resonable time frame. Lets get with it again [ H ]
 
There are two statements at the end of the review (in "Conclusion") that confused me. I'm sure it's because of my noobishness, but I'm trying to learn what to watch out for, so could someone explain what they meant by--

"The bottom PCI slot is virtually useless."

and then--

"To be fair, the layout issues are all forgivable with the exception of the clear CMOS jumper location."

I assume this is some kind of physical position problem, but what? I've gone back and read all the earlier pages, but couldn't find any mention of either of these points. All of a sudden they just appeared in the Conclusion.
 
DGig said:
There are two statements at the end of the review (in "Conclusion") that confused me. I'm sure it's because of my noobishness, but I'm trying to learn what to watch out for, so could someone explain what they meant by--

"The bottom PCI slot is virtually useless."

and then--

"To be fair, the layout issues are all forgivable with the exception of the clear CMOS jumper location."

I assume this is some kind of physical position problem, but what? I've gone back and read all the earlier pages, but couldn't find any mention of either of these points. All of a sudden they just appeared in the Conclusion.

CMOS Jumper - Out of all the places to put the CMOS jumper where it would be out of the way and easily accessible, it is placed physically on the board right underneath the 2nd x16 slot so it is hard/impossible to get at with a graphics card in there.

Bottom PCI Slot - once again, with a 2nd graphics card in there and a large cooler on the card, you can't really use that bottom slot since it is too close to the card.

There are two statements because two different people are giving their impressions :).
 
hell, I never even visit the front page anymore...I tend to find more useful information here in the forums. If I'm looking for reviews, I just google it. Then browse the first two pages that come up in the results. Some are from quack sites, but there are usually good ones in there also.

and for gods sake...stop it with the bold green print! wtf?
 
I don't understand why long lived enthusiast company like asus would do that with the cmos jumper...

you think by now they would know better.

you'd think they would have a chrome switch lever under a little piece of protective plexi or something you know .. come one .. get a clue
 
malicious said:
There's no rule against it .. don't look at if if you don't like it.


oh, no need to worry about that, I added you to the ignore list. That way I can still read the decent posts in the thread without a green mess here and there.
 
Thanks kirbyri.

In practice, is that reset jumper position just an annoyance, or does it force you to actually remove the graphics card every time you want to use it? (Is there room to fiddle around under the card?)

BTW, are second SLI slots good for anything if you're just using one graphics card? In other words, can they also be used as a regular PCIe x 1 (or x 4) slot?

And what's the deal with PCIe? Are there any rumors of things like audio cards coming out any time soon in that format?
 
In practice, is that reset jumper position just an annoyance, or does it force you to actually remove the graphics card every time you want to use it? (Is there room to fiddle around under the card?)

BTW, are second SLI slots good for anything if you're just using one graphics card? In other words, can they also be used as a regular PCIe x 1 (or x 4) slot?

And what's the deal with PCIe? Are there any rumors of things like audio cards coming out any time soon in that format?


Question 1: The CMOS jumper reset should be easy to get to. It helps as a failsafe when you push your overclock too far and the pc won't boot.

Question 2: Not sure if you can use PCIe for a regular pci slot .. I'd like to know that myself

Question 3: The deal with PCIe is more bandwidth than AGP and the ability to do SLI (Scaleable Link Interface). However, the evolution to PCIe was not brought on by necessity. An AGP bus is not at a bandwidth limit for current cards .. even the fastest ones. It was more marketing than anything else but now it looks like AGP is going the way of the dinosaur. If you are only running one card I would recommend just getting the single slot mobo. Anything in the Nforce4 Ultra chipset will do but DFI and ASUS are standouts.

If you would like.. give me your budget and i'll recommend a whole setup... best bang for your buck and all ..

mal
 
malicious said:



Question 1: The CMOS jumper reset should be easy to get to. It helps as a failsafe when you push your overclock too far and the pc won't boot.

Question 2: Not sure if you can use PCIe for a regular pci slot .. I'd like to know that myself

Question 3: The deal with PCIe is more bandwidth than AGP and the ability to do SLI (Scaleable Link Interface). However, the evolution to PCIe was not brought on by necessity. An AGP bus is not at a bandwidth limit for current cards .. even the fastest ones. It was more marketing than anything else but now it looks like AGP is going the way of the dinosaur. If you are only running one card I would recommend just getting the single slot mobo. Anything in the Nforce4 Ultra chipset will do but DFI and ASUS are standouts.

If you would like.. give me your budget and i'll recommend a whole setup... best bang for your buck and all ..

mal

The CMOS jumper should be easy to get to. One thing you do have to understand is that they crammed as much stuff onto this board as my Tyan K8WE dual processor Opteron board has. (Voltage components and second CPU socket aside.) Asus had to make certain concessions in order to get all this to fit on a normal ATX motherboard that would fit in normal cases.

You can not install PCI cards in PCI-E slots. They are neither mechanically or electriclly compatible. (If I understand the question right)

PCI-E was brought about the necessity to replace PCI. It had nothing to do with AGP. However, they found it viable to replace AGP at the same time. We also don't know what the standards commitee had decided on either. They may have mandated that both PCI and AGP be replaced with the same bus.

In either case, there are numerous manufacturing advantages to PCI-E, and some advantages to card makers as well. SLi and Crossfire are just by products of the technology.

mi1stormilst said:
Not to be rude but certainly here to speak my mind: I have found the [ H ] reviews to become increasingly infrequent, late, and near useless. The lack of competing product comparisons is inexcusable. I find myself visiting here more out of habit then of obtaining any useful information. I hole heartedly agree with the testing methodology for GPU's but that is where I draw the line. When I am shopping for a motherboard I need to see comparisons to like products, the more the better. This board has been reviewed all over the net and I have already formed my opinion a month ago or more that upgrading from my A8N-SLI Deluxe would be near pointless. I also feel that these "PREVIEWS" are poor excuses for not getting a full review out in a resonable time frame. Lets get with it again [ H ]

I'll address this for a moment. Kyle is aware that there is a need for more content and reviews. Hence why there is two of us doing motherboard reviews instead of one. Both Jake and myself had to train and then start reviewing. The first reviews take a really long time to do. The fact is, it's harder to put these things up here than you might imagine. About 20-25 hours of work goes into each review. Though the first ones took longer. We are now getting to the point where we can get them done faster than we were in the past.
 
Thanks all. But I never got an answer to my question if it's necessary to actually remove the graphics card in order to reset on the A8N32.

"You can not install PCI cards in PCI-E slots."
Actually I meant can one of the two fancy PCIe-x16 slots (for graphics) be used to accept a regular (short) non-graphics PCIe-x1 card (like one of the new RAID cards, or maybe someday a PCIe audio card if they ever start making them).

(Hope this makes sense.)
 
DGig said:
Thanks all. But I never got an answer to my question if it's necessary to actually remove the graphics card in order to reset on the A8N32.

"You can not install PCI cards in PCI-E slots."
Actually I meant can one of the two fancy PCIe-x16 slots (for graphics) be used to accept a regular (short) non-graphics PCIe-x1 card (like one of the new RAID cards, or maybe someday a PCIe audio card if they ever start making them).

(Hope this makes sense.)
I too would like to know this, as most RAID cards need an x8 slot which is near-impossible to find on anything but an SLI or nForce Professional board.
 
DGig said:
Thanks all. But I never got an answer to my question if it's necessary to actually remove the graphics card in order to reset on the A8N32.

"You can not install PCI cards in PCI-E slots."
Actually I meant can one of the two fancy PCIe-x16 slots (for graphics) be used to accept a regular (short) non-graphics PCIe-x1 card (like one of the new RAID cards, or maybe someday a PCIe audio card if they ever start making them).

(Hope this makes sense.)

Remove Graphics card question - probably depends on the graphics card. Most boards have a tendency to reset the settings themselves if the OC doesn't work after 2 reboots or something like that, so it's not the end of the world if you can't get at it. Usually you have to reset if you're changing bios revisions and with a newer board, you might need to do it more often.

PCIe slot - AFAIK, you shouldn't have a problem using any PCIe card in a x16 slot.
 
malicious said:
I don't understand why long lived enthusiast company like asus would do that with the cmos jumper...

you think by now they would know better.

you'd think they would have a chrome switch lever under a little piece of protective plexi or something you know .. come one .. get a clue

Tyan and some other companies just have a little silver push button. Which is what I prefer to see. The CMOS location is BS, but I can understand from a design perspective. There is ALOT of components packed onto a very small board.

kirbyrj said:
Remove Graphics card question - probably depends on the graphics card. Most boards have a tendency to reset the settings themselves if the OC doesn't work after 2 reboots or something like that, so it's not the end of the world if you can't get at it. Usually you have to reset if you're changing bios revisions and with a newer board, you might need to do it more often.

PCIe slot - AFAIK, you shouldn't have a problem using any PCIe card in a x16 slot.

You are correct. Any PCI-E card will do in any PCI-E slot. There are some exceptions with boards that do dual x8 slots. Mainly due to BIOS issues though.

malicious said:
I don't understand why long lived enthusiast company like asus would do that with the cmos jumper...

you think by now they would know better.

you'd think they would have a chrome switch lever under a little piece of protective plexi or something you know .. come one .. get a clue

Tyan and some other companies just have a little silver push button. Which is what I prefer to see. The CMOS location is BS, but I can understand from a design perspective. There is ALOT of components packed onto a very small board.

kirbyrj said:
Remove Graphics card question - probably depends on the graphics card. Most boards have a tendency to reset the settings themselves if the OC doesn't work after 2 reboots or something like that, so it's not the end of the world if you can't get at it. Usually you have to reset if you're changing bios revisions and with a newer board, you might need to do it more often.

PCIe slot - AFAIK, you shouldn't have a problem using any PCIe card in a x16 slot.

You are correct. Any PCI-E card will do in any PCI-E slot. There are some exceptions with boards that do dual x8 slots. Mainly due to BIOS issues though.

kirbyrj said:
Remove Graphics card question - probably depends on the graphics card. Most boards have a tendency to reset the settings themselves if the OC doesn't work after 2 reboots or something like that, so it's not the end of the world if you can't get at it. Usually you have to reset if you're changing bios revisions and with a newer board, you might need to do it more often.

PCIe slot - AFAIK, you shouldn't have a problem using any PCIe card in a x16 slot.

I did have an issue once where I could not get it to do so. I did need to clear the CMOS manually. I would have had to remove the secondary PCI-E card if I had one installed. Any card would get in the way. The jumper is located under the PCI-E retention clip.

DGig said:
There are two statements at the end of the review (in "Conclusion") that confused me. I'm sure it's because of my noobishness, but I'm trying to learn what to watch out for, so could someone explain what they meant by--

"The bottom PCI slot is virtually useless."

and then--

"To be fair, the layout issues are all forgivable with the exception of the clear CMOS jumper location."

I assume this is some kind of physical position problem, but what? I've gone back and read all the earlier pages, but couldn't find any mention of either of these points. All of a sudden they just appeared in the Conclusion.

Kyle and I both made mention of the layout issues including the clear CMOS jumper. Though I didn't bring it up in the layout part of the article. I covered the other layout issues mainly. It wasn't a conscious decision to omit that from the layout portion. I wrote that part of the article before actually needing to use the rest jumper.

See I write most of the article as I go along. So I handn't really acknowledged the issue yet. I noticed it, and thought that it was minor since a hotkey or two unsuccessful boot-ups would reset the BIOS to defaults if needed.

It just didn't happen. I do the overclocking last. So if I break anything I don't have a delay in getting the rest of the review finished.
 
malicious said:
WTF is THAT shit?
I mean, I turn my head away from [H] for 5 minutes....
now there is a giant infomercial in my face at all times.
My sig is mutilated, everything in general is off center and now people are deciding to type in giant colored letters that make my eyes bleed.
Truly wtf is that shit.
 
I am very disappointed in the way the A8N32 SLI is given Kudos for performing higher than the rest of the boards when in all the tests it is ran with the top of the line processors and not compared to the others with the same one. I feel these tests should have been done with the same processor on every board. Of course a FX57 and X2 4800 will perform better than the 3500 and 4000. Also according to the setup specs the A8N32-SLI used 2 gigs of memory where the other systems used one.
1GB (2x1024MB) Corsair XMS DDR 4400UL DDR550 (2.5,3,3,5)
Should read
2GB (2x1024MB) Corsair XMS DDR 4400UL DDR550 (2.5,3,3,5)
 
bigapl said:
I am very disappointed in the way the A8N32 SLI is given Kudos for performing higher than the rest of the boards when in all the tests it is ran with the top of the line processors and not compared to the others with the same one. I feel these tests should have been done with the same processor on every board. Of course a FX57 and X2 4800 will perform better than the 3500 and 4000. Also according to the setup specs the A8N32-SLI used 2 gigs of memory where the other systems used one.
1GB (2x1024MB) Corsair XMS DDR 4400UL DDR550 (2.5,3,3,5)
Should read
2GB (2x1024MB) Corsair XMS DDR 4400UL DDR550 (2.5,3,3,5)

Well not all the boards reviewed were done at the same time. There isn't always the luxery of comparing exact processors on every board. In many of the tests, it wouldn't even matter. Obivously in game and some app tests yes. The Oc'ing potential of the board is discovered through FSB overclocking. Often by reducing the multiplier of the CPU in order to ensure that the board is what's being tested. Not the CPU. In those cases, the CPU makes little difference.

Those other boards are there for reference. Not for direct comparison. We aren't doing a total motherboard round up comparing each board to one another with all the same hardware. We've never done it that way. It shouldn't be interpreted as that. Also, when comparing boards from any vendor, there is usually not more than a 1% or 3% variance from one board to the next. Where boards really differ is in quality and stability, and feature set. Not in actual speed between one another.

I made an error on what was quoted above. I am surprised that my mistake wasn't caught in the editing process. They usually are.
 
"The A8N32-SLI dominates the competition in this game.. "
"Again, the A8N32-SLI outperforms its competitors. '
"While the A8N32-SLI shows a massive lead over all the other boards in Doom 3, these results should be kept in perspective. The performance advantage enjoyed by the A8N32-SLI is probably due, in part, to driver changes made between reviews. "
"The A8N32-SLI takes top spot in this test. "
These are more general statements and not actual results seeing as the testing is done with excesssive variables.
 
nst6563 said:
oh, no need to worry about that, I added you to the ignore list. That way I can still read the decent posts in the thread without a green mess here and there.
Unfortunately that doesn't work because of how many people are quoting him....ah well
 
bigapl said:
"The A8N32-SLI dominates the competition in this game.. "
"Again, the A8N32-SLI outperforms its competitors. '
"While the A8N32-SLI shows a massive lead over all the other boards in Doom 3, these results should be kept in perspective. The performance advantage enjoyed by the A8N32-SLI is probably due, in part, to driver changes made between reviews. "
"The A8N32-SLI takes top spot in this test. "
These are more general statements and not actual results seeing as the testing is done with excesssive variables.

Does it really now? (-;
 
I don't know if the bottom slot is useless (I have my X-Fi Elite Pro card in there), but it IS a pain in the ass.

Under that slot are the USB and Firewire headers. If you have the board mounted, and a card installed here, in a typical case, it is impossible to get to those connectors.

You have to pull the card to get to them.

-Larry

DGig said:
There are two statements at the end of the review (in "Conclusion") that confused me. I'm sure it's because of my noobishness, but I'm trying to learn what to watch out for, so could someone explain what they meant by--

"The bottom PCI slot is virtually useless."

and then--

"To be fair, the layout issues are all forgivable with the exception of the clear CMOS jumper location."

I assume this is some kind of physical position problem, but what? I've gone back and read all the earlier pages, but couldn't find any mention of either of these points. All of a sudden they just appeared in the Conclusion.
 
Yeah, something is broke...

-Larry

qbackin said:
I mean, I turn my head away from [H] for 5 minutes....
now there is a giant infomercial in my face at all times.
My sig is mutilated, everything in general is off center and now people are deciding to type in giant colored letters that make my eyes bleed.
Truly wtf is that shit.
 
The point was that when the second video card is installed, the bottom PCI slot is useless if you have a dual slot cooled video card. Obviously a single slot card won't have that issue, though airflow is restricted some.

Plus you do have issues getting to CMOS and firewire/USB headers.
 
sac_tagg said:
I too would like to know this, as most RAID cards need an x8 slot which is near-impossible to find on anything but an SLI or nForce Professional board.

The A8N32 works perfectly with the Areca 1220 in the bottom PCI-e slot. Aside from a bad drive that caused me about a day of grief, the card and array has been stable for about a week. I would venture to guess that it would probably work in the other slot, and I havn't attempted to reset the CMOS but the board is very tight overall in this configuration.
 
bigapl said:
I am very disappointed in the way the A8N32 SLI is given Kudos for performing higher than the rest of the boards when in all the tests it is ran with the top of the line processors and not compared to the others with the same one. I feel these tests should have been done with the same processor on every board. Of course a FX57 and X2 4800 will perform better than the 3500 and 4000. Also according to the setup specs the A8N32-SLI used 2 gigs of memory where the other systems used one.
1GB (2x1024MB) Corsair XMS DDR 4400UL DDR550 (2.5,3,3,5)
Should read
2GB (2x1024MB) Corsair XMS DDR 4400UL DDR550 (2.5,3,3,5)


That's the whole point of this thread.

Why do the [H[ard reviews have confounding varialbes?

not bashing or flameing ... just wondering


hey look no green --- i'm cured ;)
 
malicious said:
That's the whole point of this thread.

Why do the [H[ard reviews have confounding varialbes?

not bashing or flameing ... just wondering


hey look no green --- i'm cured ;)

Well there a few reasons why. The main one is we don't benchmark multiple boards against each other. We aren't doing motherboard roundups here. We do each review one at a time. I have to use previous reviews for comparison purposes. As such, if a significant amount of time occurs between motherboards compared in some instances. I don't have 2 or 4 reviews using the same processors to compare with. I also won't compare stats for reviews not published. That right there leaves a gap until a few reviews are done and published before we can compare apples to apples.

With motherboards I agree that apples to apples would be nice, its just not always possible. So when I choose which reviews get compared, I try to get them as close as possible. This just happened to be one of if not the first FX-57 reviews we've done. So naturally it will boast some higher numbers than other earlier reviews.

Still much of the tests for drive speed, network speed and others are definately comparable. The CPU has nothing to do with those tests.
 
Back
Top