Discussion in 'Article Discussion' started by WilHarris, 20 Dec 2005.
Went to buy one of those last week but they hadn't got any in stock so i got the A8N SLI Premium instead - patience is not one of my virtues, especially when i've just walked several miles to the computer shop. Ah well.
I've not used the Premium, but if it is as good as the A8N-SLI Deluxe, it'll be a pretty good motherboard.
Anyone knows a mobomaker with the same specs but with a better audio chip? I curently have a two a8n-Deluxe boards. The sound is so damn bad. The autosensing senses anything i have NOT connected. Even trying to manualy change the inputs/outputs are a disaster. If i play a audio file with a "spike" in it. The audio volume drops to zero and takes 5 minutes to get back to normal volume levels. I can not disable the *CPU FAN failure* warning either (i have wc cooling)
The other board has a defective casefanheader. And it takes 10 minutes to boot from start to postscreen. I tried different biosses to fix this. It has proberly something to do with the defective casefan header...
pfff.... sorry for getting offtopic.
The MSI K8N Diamond (nForce4 SLI) and K8N Diamond Plus (nForce4 SLI X16) use Creative audio chips onboard. Also, ATI's Xpress200 chipset has the Realtek ALC880 High-Definition audio chip onboard.
same about 2 weeks ago, just needed the comp there and then. also bought the premium, cant decide wether i shuld upgrade or not, after al, it was what i originally planned to have in my pc
You not been speaking to mister_tad lately?
He's having a wonderous time with his Premium!
Would this be the mobo that you'd build your system around.... bigz?
If I was building an SLI system without any add-in PCI cards... pretty much definitely, yes.
A bit of a surprise that the Asus' boys didn't add crash free bios to this board too, since the A8N-SLI Premium has it (didn't bother to check does the deluxe have it).
I've had an order in for the 32-SLI for quite some time now but because I had to send my DFI LP SLI-D to the doctor (problems with pci slots), I didn't have time to wait for 32-SLI and so I got the Premium straight of the shelf of a local store (was a bit lucky there since it was the last they had at the time).
I'm somewhat dissapointed about the lack of tests. Doom3 and FarCry aren't exactly the most demanding games anymore. Since one of the main selling points for this mobo are the added SLI lanes, I would have expected atleast a test with F.E.A.R.. I'm sure nobody would have minded to see some other games tested too to clarify the difference the added SLI lanes make, or do they.
Maybe the lack of game tests has to do with BT's method of testing motherboards, but in this case it's not exactly sufficient. I think that testing methods that miss important features aren't exactly what would be expected - which in BT's case, for me anyway, is alot.
One other imporant thing is the 8-phase power...it was hardly mentioned, so how are we supposed to understand that, it's not any good? didn't seem to affect anything or that you didn't know how to test and figure out if it does anything useful?
Do I detect some rushing in the publishing of this review or...is it just my imagination? somehow seems a bit...unfinished. I'm not too happy to be 'complaining' like this because I've come to appreciate the way you look at things here at BT, but that is also the reason why I had to say all of this.
Or those who won't be using SLI, call me insane for spending £150 on an SLI mobo while not intending to buy a second graphics card, but I wanted a mobo that would overclock very well, while not being as unstable and unpredictable as a DFI, this board seems to be that board.
With only one graphics card, I can use 3 mobo slots, although, maybe manufacturers aren't pushing an excess of PCI slots on motherboards like this, because they're aware that many who'd care to fill them, either have the know how, or already do, have a second PC/server.
I don't know about you, but I'm just assuming that the extra PCI-E lanes for the graphics cards are, for now at least, just a nice sounding feature, and not something thats going to help a whole lot for another few generations of cards.
Question Bigz, this isn't entirely important to me anymore, but how much of a difference if any at all, would using a 20pin ATX connector in the 24pin ATX socket make? Not safe at all, decreased voltage stability, or little difference?
It does, I missed it out of the overview by mistake. It's added in there now.
As I've said later in this post, I played around with the board for a week prior to starting the review on it. That involved playing games, using SLI. To be honest, I didn't find it any different to a normal nForce4 SLI - the 'highest playable' settings were the same in both cases. SLIAA performance is better, but SLIAA is one of those things that works sometimes, but not on other occasions.
It's certainly not something I'd recommend using in its current state - it's also only really viable in older games and I would only use 16xSLIAA, as I don't think that 8xSLIAA is worth the performance hit when 8xSAA is as good (if not better, because it's 4xMSAA with 2xSSAA applied on top rather than two lots of 4xMSAA blended together to make 8xAA - think temporalAA) with generally better performance.
Maybe the games should be updated, but there's the time factor that plays into this - more games have been tested, but the results are not comparable with anything at the moment because no other boards have been tested on those games. It's not just as easy as firing up another game on any particular board and recording results, because I prefer to run things in a set order with the system state being exactly the same when the tests are run on each particular motherboard. I install things in the same order on each board, and it'd require more work than there is time for. I'll further the reasoning for this later in my post.
It generally means the power is much cleaner when it gets to the CPU, however, there are very limited ways to test whether this is the case (or not). The board seemed to overclock better than the A8N-SLI, which gives a reasonably good indication that the power circuitry is doing its job.
The motherboard review certainly wasn't rushed - it had as much time spent on it as would be spent on any other motherboard review. I'd messed around with the board for about a week prior to starting the review on Thursday afternoon, and was finished yesterday evening with stress testing done over the weekend.
Unfortunately, Rich is now not reviewing motherboards anymore due to commitments with University that leave him no time to work on motherboards. That means that the review hasn't been shadowed - it was done straight out of the box by me. This is slightly different to the reviews that Rich did, which were shadowed by me. I used to spend a good amount of time working with him to make sure that the reviews were as good as could be with the time and resources available.
To further the arguement that the review was 'rushed', I'll be lucky if I get more than 2 days off for Christmas this year with a product launch happening on the 27th - I've only just got the kit for that today. Tomorrow is a bit of a Christmas get together with Wil, Rich and some of the guys from TR. That leaves me with Thursday, Friday, and Saturday ready for publication on Tuesday (surprisingly, I have no plans to work on Sunday or Monday). I don't think you understand just how much work goes on behind the scenes - I would guess you'd be surprised if me, Wil and Geoff rack up close to 300 hours a week between us, pretty much every week, without fail.
Based on my experiences with the board, that is what they are in pretty much every scenario that the board went through in my 1-week FAMEX before starting the review.
In all honesty, I don't know - the more wires, the more clean power can get to your board. I'd recommend using a 24-pin one as soon as you can afford it (and if your board supports it, etc), but I used a 20-pin PSU on the nF4 SLI-DR for a long time, as I lost my 20-pin to 24-pin adapter for the OCZ PowerStream. I've found it since then, though.
Using a 24pin, as is using the additional molex plug, an advantage, because you spread the power use out over more of the motherboard minimising hotspots as well as minimising extremes in electric field and subsequent crosstalk across what is an extremely small lateral space. If you're sticking your few hundred watts down only twenty four wires (20+4pin 12V) as opposed to 24+4+4 there's going to be a stability difference in how power is accessed across the board.
Having said that, ALL the tests I ever did were with a 20pin, 380W PSU and I never ever had a problem with stability. Using SLI I would always plug in the 4pin molex, simply because it was recommended in the manual to do so.
Every single board should work with both, unless it specifically requires a 24pin by the manufacturer.
I could tell you without even having used the x16 board there's FA difference between x8 and x16 at this time. Even using a single graphics card in x8 and comparing that to the same at x16 there's literally no performance difference so I personally wouldnt bother with x16 if I was looking for a performance increase. It's hard to test sheer bandwidth that isnt either limited by CPU or graphics card processing power. If you are arguing high detail is more intensive a test, then you are CPU limited, so that doesnt help either.
Im supprised NV hasnt STILL yet updated their onboard audio. Even though they have an Intel licence, it crazies me out.
Great review Tim; I think the only issue people are really observing is the stylistic difference between you and Rich. Very straightforward, and it illustrates the points you mean to make well.
I appreciate your efforts to find issues with this board as well; it gives potential buyers more of an idea of what to look for, and gives me more ways to make my Ultimate Motherboard better than this one
Should've had a look online...
Have a little question on heatpipe on this mobo, would it work if the mobo is mount inverted in case like lian-li vseries same as btx format...
Nope. Aside from airflow concerns for the radiator fins on the end of it, heatpipes tend to be unidirectional because they function on the idea of evaporative cooling.
It might work minorly simply due to all the copper, but the effectiveness would be severely hindered.
Thus the comments about the reset cmos jumper aren't exactly right, right?
That's good to know, wouldn't have had said anything about it if there had been some sort of side note in the review itself about it though =]
I'm aware of what it's supposed to do but not mentioning about it made me wonder. I'm pretty sure you're right about the overclocking indicating that it's working, I guess I'm just not jumping into such conclusions when I'm expected to. Somehow thought that there might be some other desing related points too, that might improve such results. And so, it wouldn't have hurt to say something about at some point.
Ok, my bad there, sorry.
I'm not exactly sure how I should take this...but let me just say that I appreciate this site and _most_ of it's editors alot and BT is at the top of my reading list every time there's a new review / article / and all.
Your guess is quite accurate, since I didn't really think you spend such vast amouns of time on providing us with new things to read about. Although I rough idea because of reading about your testing methods etc.
Like I said, I try not to critisize too much and I really don't mean to offend anyone by doing so. Thanks for the article and for clearing things up.
It's said to have a CrashFree BIOS, but if that means it boots when settings are 'too much' for the board/CPU/memory to handle, it simply doesn't work.
There were several occasions - when overclocking - where I had to reset the CMOS jumper. These occasions were where I was a little optimistic about how far I thought the board (and the relatively poor memory controller on our FX-57) would overclock.
If something was 'rushed' I wouldn't defend my position quite so profusely.
We try to be very open with you guys and we welcome criticism as long as it's constructive. At the end of the day, we want to provide you with the information that you want about a product in a clear, thorough, easy to read and concise manner. We're trying to work on ways of making the content as efficient as possible so that we can bring you more articles each week.
Your comments have been taken on board. I guess that the game tests came up a little short of expectations in this review and we'll try to rectify that in future motherboard reviews.
Separate names with a comma.