Printable View
Correct. The reference frequency tells the MCH what speed it should assume it's running at and to set its internal timings appropriately. The lower it's set, the lower the MCH latencies. Just like memory, there's a tradeoff between latency, speed, and voltage.Quote:
Originally Posted by xMrBunglex
ok, that makes sense, thank you.Quote:
Originally Posted by gtj
And so along that same line of thinking can I assume that 1:2 in 200 will hit a wall before 266, and 266 before 333, etc, as FSB increases? Is this overcome through increased vMCH?
and this all more or less goes back to my issue of not being able to effectively set my multi below 10, and therefore i am trying to maximize bandwidth in other ways.
A little bit can be overcome by voltage but not much. It's mostly FSB.Quote:
Originally Posted by Sean-E-Boy
You should report an issue to Intel on the multiplier issue. I'm going to report it as well.
http://supportmail.intel.com/scripts...d=36,2778,2807
issue reported.
and escalated
well, i dont think that my strategy is going to work:(
it seems that 200/800 tops out around 330FSB for me and 266/1066 is about the same. going to try a few other things, but it is probably back to 333/1333 for me until i can get this muliplier thing worked out.
anyone had any luck with GTL values and independent core stabilization?
Sean,
Have you tried enabling EIST under the power menu to see if that makes the multiplier setting work. It works for me if EIST is enabled.
yes, i tried that too...no workie.
i really don't know much about GTL values. i'd like to learn more though.
attached is the reply from Intel. I expected something like this, and hopefully they still keep track of the issue and pass it on to someone that can do something about it.Quote:
Originally Posted by Sean-E-Boy
However, I didn't expect that more than half of the reply would be about "please rate our reporting system..."
I have some really good documents related to the issue (but i hope that you have a PhD in electical engineering!). I will post them later today once I get home.Quote:
Originally Posted by xMrBunglex
as promised
<<attached>>
the intel one is more technical than the other
sean
Don't get caught up in the "overclocking" spiel. Reply back with a statement that "A BIOS feature is not working correctly". That should key them to pass the case on to the BIOS team. It takes a little patience (just like the board), but you CAN get to a live person who actually has a board in front of them and will try it out. They've gotten a little more "enthusiast friendly" but the autoresponder and tier 1 hasn't got a clue.Quote:
Originally Posted by Sean-E-Boy
Good finds. :)Quote:
Originally Posted by Sean-E-Boy
Thank you.Quote:
Originally Posted by gtj
And as per your recommendation, I have replied to Intel's reply.
Typical Intel reply, always oc is not supported, even if the matter is not directly related to oc. I guess it's always their first move to see if we just let it go...
Anyone knows about the C2 state option? I guess it's a new processor state but any details? C0 and C1 are known, this one must be new.
IS this board a good Overclocker With X3360 (Q9550) CPU ?
What would your goals be?Quote:
Originally Posted by X2beach
Has anyone tried a QX9770 on the DX38BT, since the bios for the X38 is currently the same as the X48 it should work right? And by working i mean will it detect the proper speeds automatically.
Just curious.
does this board allow you to tweak the GTL and anyone got q9550 to run 450 bus on this board? DFI I cannot get it to run more then 442bus
The motherboard will obviously run at a 1600 but whether it will be automatically detected I'm not sure.Quote:
Originally Posted by rchelicopter
The latest BIOS does allow you to tweak the GTL but I had an E6850 stable at 450x9 without GTL adjustments. Can't say on a Q9550 though.Quote:
Originally Posted by X2beach
I am thinking about getting the DX48BT2 since I have some OCZ DDr3-1600 just laying here I like to use 4gigs of it..
I wonder if this board is more stable the the DFI X48 ?
Quick update on my set-up.
I got some time to put everything on the bench and run memtest. It looks like I can get all 4GB running 1333 stable but I have to set the MCH @ 1.4.
Anything I should be concerned about running it that high ?
you'd actually be willing to buy a new motherboard just so you can get an extra 8 MHz FSB? you obviously have too much money - i'll be glad to take some off your hands :D i accept cash or PayPal.
the highest i ever tried to get my BT was 467 FSB (for an even 4.2 CPU) and it handled it easily with moderate volts and stock NB cooling with no fan. a former member of this thread hit 500 FSB (i think he's the only one who tried it). none of us have an X3360 though. i'm not sure how that would affect the results.
I got one on the way and the Asus Striker2 Extreme.
I will post results from both this weekend..
i've been running my vMCH at 1.425 and vFSB at 1.35 since i got the board. no NB fan, stock thermal pad on the heatsink... totally stock. my NB idles at 30° and i think i saw it as high as 43° once during an Orthos run in warm weather. you'll be fine at 1.4.Quote:
Originally Posted by Randy
Thanks for the info. I think I will try a little more and see if I can get this memory up some. I'm running Vista 64 on my Phenom machine with 8GB DDR2, and I am really pleased with the 64bit overall. Crossfire still blows on it just like it did on this board for me anyways(the Intel machine spoiled me). 4GB is essential with the 64bit IMO.Quote:
Originally Posted by xMrBunglex
I was planning on parting the Intel machine out, but I just can't bring myself to stop playing around with it.
Anybody here go from an 8800GTX to a pair of 3870's crossfired that can shed some light on the performance differences. Yeah, I combed the net for comparisons, but every performance site has thier own opinions and testing platforms, so you really compare them on 38BT.
To completely understanding the reference voltages is complicated. But a basic understanding of what to do with them is pretty easy.
Adjust it in one direction until you get a failure. The failure criteria can be anything, as long as it's consistent. Booting to Windows is a pretty good test. Fine tune it down to one step = pass, the next step = fail. Then adjust it in the other direction and fine tune in again. You'll end up with a min and max number, usually referred to as you margin window. Then you simply pick the number in the middle of the window, and set your final value at that middle point. That will be your best stability point, at the voltages and frequencies that you tested at.
You don't need to do long term stress testing to find the absolute min and max stability points, your not going to be running out there, you just want to use a consistent, and quick test to fine min and max, so you can then calculate the middle point.
As you increase frequency the min and max numbers, your margin window, will shrink down. The min might shrink faster than the max, so you might start failing because your negative margin, adjustment range below default, has collapsed down to nothing. But you might still have positive margin, adjustment range above default. You could raise your reference level to recenter in the margin window that is remaining.
If you jump too far on the adjustment and cause a hard no boot failure, you'll have to pull power from your power supply and move the BIOS jumper to get back going again, so go in small steps :)
thank you very much, Anubi. i've read about the theory behind GTL before but that was the first practical advice i've seen on how to optimize them. i appreciate it.
Hello people :)
I've got this board since a couple of weeks but there is a big issues: I can't overlock it !
I if change any fréquency (RAM, FSB or other), the board loop boot ... It's a lil' bit frustrating ,and I've try with 2 differents CPUs then ... I guess it's the board.
If you guys have any suggestions ... it would be great !
Thanks ;)
Herwawan.
(Used system: E8500, DX48BT2, 2Go OCZ DDR3 1600 XMP, 80Go SATA/Lecteur DVD SATA, 9800GX2, Enermax Galaxy DXX 1000W.)
that's the watchdog. move the CMOS jumper to enter maintenance mode and turn the Watchdog off. it will stop looping after that. don't ask why Intel included the watchdog with their enthusiast motherboard - none of us know.
:o :o :o !
It finaly WORKS !
xMrBunglex, I love you :D
More seriously, yeah it works and it's good for me huhu
There's still a (big) problem: the multi change all alone :D I've disable EIST and C2State, but I've style can't control the multi by myself. I wanted to have 3.6Ghz, then 9x400, but it set 9.5 for the max limit, and the EIST seems to be still enable (it change the multi in idle etc ...)
What's wrong ?!?
that is C1E, which cannot be disabled in BIOS. again, we don't know why. you can use Everest to turn C1E off if you want, but i'm running a pretty high overclock on mine and the C1E hasn't caused me any stability issues at all.
mmmh Ok but in Everest (or OCCT by the way ...) I've got this error: Invalid Floating Point Operation ... What does it mean ?
are you running a stability test? some of us had to up our memory voltage a bit to make it stable at default speed and timings. i don't know if the BT has vdroop to the DIMMs or what. my memory is rated at 1.8V and i had to set it at 1.9V to make it stable at its advertised speed and timings.
Yep I use CPU Stress MT, and it says : It's ok dude your cpu at 4 gig' run fine :D
Because of this stupid multi, I've set 425FSB, so 850Mhz (cas 8) for my ram, and 4(,035)ghz for the proc' ...
Last question: I own OCZ XMP 1600 Modules, and I don't find good timing for making them work at 850 with cas 7-6-6 ... Do you know good timings for X48 ?
i don't think my memory is quite as good as yours, so it would be an apples to oranges comparison. also, i'm still refining my overclock. i haven't quite gotten everything finalized yet. what i have so far is:
FSB @ 440 (333/1333 reference clocks for a 1:2 ratio)
vFSB 1.35, vMCH 1.425
memory @ 880 (DDR 1760) @ 1.98 vDIMM
8-8-8-24 2T 88-5-12-6-6 timings
i'm able to pass Memtest with tighter timings (7-7-7-18 2T 45-4-10-5-5), but i fail Orthos in less than a minute. that's why i had to loosen my timings.
GTJ has OCZ memory and might be able to help you more than i could.
after i get all my frequencies and volts right where i want them i plan on optimizing my GTL settings. i'll share my settings with this thread after i've done all the work.
i should also add that i haven't found any benefit to a 1:2 ratio yet outside of synthetic benchmarks. my Everest and SiSandra benchmark scores are very impressive, but i'm still compiling video at the same speed as i was with my memory at DDR 1500 and slightly tighter timings. at least Adobe Premiere Pro and ConvertXtoDVD 3 don't seem to benefit from my increased memory bandwidth at all.
Ahah ok :) but so far I still can't load any benchmark like everest or sandra: "Invalid Floating point operation" :'( In the worse case, the bench just don't move :D
oh, sorry. i thought the error you got was during stability testing. i assume you've already passed a stability test? also, it would probably help if you read all the way through this thread to make sure you understand how screwed up the chipset drivers are. make sure you have the latest drivers (not the ones Intel's website call the latest). if you've already done all of the above, i don't know what to say.
Try putting everything back at stock and see if your errors go away. It is very possible that your overclock allows you to boot to Windows (or your OS),Quote:
Originally Posted by herwawan
but the errors may indicating that your hardware is unstable at those settings.
For example, I can boot to Windows with my machine overclocked much higher, but random file errors and system reboots occur running programs. Since I want to do more with my machine than sit and look at the desktop, I had to reduce the overclock and adjust it until it is stable. You can verify stability by running prime95 or one of the many other system stressing utilities.
Ok, I've got these errors at stock speed too :D
I use CPU Stress MT to verify stability, and prime ...
I'll try find others drivers, and we'll see after
EDIT: I've got the 8.3.1.1010 driver, and when I try to update it said: you've got the last one you dumb why wan't to come back ?!?
Ahaha Anyway the "Intel Desktop Control Center" don't work too, I guess for the same reason as the others tools ...
What's wrong ? I'm using Windows XP SP3 ..
Edit 2: Ok i've find infinst_autol_8311011.exe and I've udpate it ... but there's still this stupid error :( Anyway there's a good news: THE EIST IS NOW DISABLE !!
here is the link: http://mainboard-drivers-and-utiliti...(x32-x64).html
I'll check for the multi now :) ... and the floating point error :o
i know it's totally unnecessary, but i bought a Scythe Mini Kaze 40mm fan for my north bridge. i bought it because it was only $4 and i was bored. my case has a side fan that already blew a fair amount of air on the NB anyway, so i was curious to see if the Scythe had any affect at all.
http://i21.photobucket.com/albums/b2...ve01/NBFan.jpg
the fan is totally silent as far as i'm concerned. i can't hear anything. it doesn't have any affect when i'm idle, but it did a pretty good job at full load. i ran Orthos for 30 minutes in a warm 75°F room and my NB only went up by 2°C according to Everest 4.50.
How are you measuring your NB temps? The 3 temp sensors are CPU, Motherboard (which is the 2 o'clock position relative the CPU) and DIMM which is between the 2 sets of DIMM slots.Quote:
Originally Posted by xMrBunglex
i don't have an infrared thermometer or anything. i'm just using Everest Ultimate 4.50 and going off the Motherboard Temp. i know it isn't perfect, but my Motherboard Temp would sometimes climb as high as 10°C over the idle temp when i ran Orthos in a warm room (i live in Southern California). the reading barely moved when i ran Orthos with the fan installed.
i've done some more testing and i honestly don't see any benefit to running my FSB/memory at a 1:2 ratio. i've run CineBench, SuperPI, 3DMark, and i've compiled hours of real video. the only benchmarks that run better with a 1:2 ratio are synthetic memory tests like the ones found in Everest and SiSandra. i can't find any real world benefit though. my CineBench score is actually a little better with my memory at DDR 1480 7-7-7-18 1T than it is with my memory at DDR 1780 8-8-8-24 2T.
food for thought. i welcome any input from others that have done testing. until i hear something new, i'm sticking with the lower ratio and tighter timings.
EDIT:
i just converted 970 MB (in 216 files) of .MP3's with a mixed bitrate (160 Kb/s and 192 Kb/s) into 128 Kb/s .WMA files using Windows Audio Converter. i did it with my memory at DDR 1480 and DDR 1780 using the above timings. both conversion runs were within a second of each other @ 13 minutes, 37 seconds.
maybe quad core CPU's benefit from the extra memory bandwidth more? i just don't think my CPU can use all the bandwidth my memory is capable of providing.
Long time no new BIOS releases from Intel for the BT, either they believe it's perfect already and something big is coming...
Only new thing out latelly was IDCC 4.1.0.128 beta (still) with 64bit support and performance tools, finally.
I'm still waiting for the performance level (or tRD) to be available in BIOS, and there's also C1E missing still.
What I would really like to see for this board would be a version of UEFI instead of the good old BIOS, the board supports it and now, since SP1, so does Vista (only 64bit) but since Intel was the big propeller why still no such thing for the desktop yet? Others (MSI) are already sampling. Can't wait to see it on my BT, if ever...
is anyone running the latest IDCC? i'm afraid of it since i found out it was causing all my crashes back when i got this board. i love Everest and am not willing to part with it, so i won't reinstall IDCC until it plays nice with other monitoring programs.
MB
v4.1.0.25 is working fine for me, haven't tried v4.1.0.128 yet. IDCC is the only one I'm using so I can't say if it behaves with others around.
where are you seeing 64-bit support? i still only see 32.Quote:
Originally Posted by Manel
Sean
Thw initial page was not updated, you need to enter the download page, agree to the terms and then:
Download the Intel® Desktop Control Center v4.1 beta for Intel(r) Desktop Boards DX38BT and DX48BT2:
Intel(R) Desktop Control Center 4.1.0.128 beta (12.2MB) for both XP (32 & 64bit) and Vista (32 & 64bit)
IMPORTANT: click here to view installation instructions and release notes
Passmark BurnIn Test 64bit (4.42MB)
Passmark Performance Test 64bit (4.88MB)
Doesn't work for me either. I had to go back to 4.0.1.25. Luckily I hadn't deleted it yet.Quote:
Originally Posted by Manel
Hi...
I found mine would loop boot also but only 3 -4 times then fire up as normal after changing memory settingsQuote:
Originally Posted by herwawan
MrBunglex... I've noticed from your rig shot you are running two crossfire straps. Have you tried just running with the one strap, the one closest to the outputs (furthest away from the power connectors)?
It made a huge difference on my rig.
i haven't tried running one CF bridge, but my CF performance is great. i've heard of people having problems when they have one bad bridge. that might be the case with you.
Just thought I'd ask as it's a recommendation from ATI forum :shrug:
oh. i put two bridges on as soon as i installed my cards and i haven't removed them. i've never tried it with a single bridge, but i couldn't imagine it working better. the ATI forum is mostly where people go when they have questions - there aren't a whole lot of answers to be found over there. this forum is much better for such things.
Does anyone here know if you can do a 2 drive RAID 0 array and a separate RAID 1 array at the same time on the Intel SATA controller ?
Any thoughts on issues ?
So 4 drives - 2 RAID 0 and 2 RAID 1.
Third time my Terrastation had to resynch my RAID 5 array over the past 2 years and I am getting nervous.
I dont see why not, i have 1 2-drive 0 array, 1 3-drive array and 1 non-raid drive all running together. even with lots of simultaneous transfers between all 3 sets i dont see any performance drop, so bandwidth should be fine, and i hope with this many sata ports on the same controller that it could handle 0 and 1 at the same time...seems like a waste if it couldnt.Quote:
Originally Posted by Randy
I've got a 2 drive RAID-0 and a 4 drive RAID-10. No issues at all.Quote:
Originally Posted by Randy
Good to know.Quote:
Originally Posted by gtj
would this be a good board for a Q9450?
would like to overclock a decent amount, maybe around 3.6GHz max
you only need a FSB of 450 to hit 3.6 with that CPU, which the Bone Trail can do easily. a lot of motherboards can do it. do you need CrossFire? if not, a P35 motherboard will do the same things and cost less. if you do want CrossFire, there are other mobos that work well too. i've been eyeballing DFI boards lately. i haven't found any obvious benefits to DDR3 memory either.Quote:
Originally Posted by Ender17
to be perfectly honest, i wouldn't recommend this mobo to a friend. i've learned to live with mine, but i miss my old Asus.
that's all I wantQuote:
Originally Posted by xMrBunglex
would prefer a DDR2 board, only using a single GPU - 8800 GTS 512
so many options, I don't know what to get :confused:
Well, 450FSB seems to be the sweet spot for this board and with a Q9450 that puts you right at 3.6GHz. Use DDR3-1800 memory and you're set.Quote:
Originally Posted by Ender17
BUT... I think everyone here will agree that this board takes a lot of time and care to tune. It's not for the casual overclocker or someone who doesn't have the time and patience to put into it.
I'm out then. Just want something simple. Suggestions for other boards? DDR2 preferably.Quote:
Originally Posted by gtj
Asus P5K-E is always a good choice.
the ip35 pro is great board!!!!!!!!!!!!!!!!!!!!!!!
Can anyone recommend sub timings for my memory (check sig), I'm currently using Bios 1478 but would like to start using ver 1521. The problem is I'm unable to get a stable overclock with 1521 and I'm leaning towards the problem being the memory sub timings.. Thanks
i have mine at 45-4-10-5-5 1T. when i overclock the RAM i use:
88-5-12-6-6 2T
download Everest Ultimate and go to the Motherboard/SPD tab and it will show your memory's SPD and XMP profiles. it has complete timing information. as far as the sub timings go, write down the 2nd, 4th, 5th, 6th, and 7th numbers. those will be the ones you can enter in your BIOS.
Since my fancy Tax Rebate arrived, I'm heavily thinking about a new build... Anyone already tired of their DX48BT2, yet? PM me. :)
-bZj
Thanks for the information, I tried entering those "sub" timings into Bios 1521 and got a little further than I did before as it had tighter timings by default.. But i still need to do a little work to get it as stable as it is with Bios 1478.Quote:
Originally Posted by xMrBunglex
On a side note can anyone on this forum that is using watercoolin tell me what the "normal" difference between the waterblock temp and the cpu be? Mine is about 10c higher at idle and 25c to 30c difference at load.
The thermocouple i have mounted on the side of the waterblock reads 30c and the cpu reads 40c in everest, under load the block is about 45c and the cpu is about 60c is this normal or is my waterblock inferior.
Long time no BIOS updates from Intel, 1521 seems to be the end of the line here for us. Either there's no more updates or they preparing something really big.
I wish this time they would release a version of the Framework (UEFI) instead of BIOS for this mobo and leave god forsaken legacy code for good. Dream on...
Also, IDCC will never leave beta status I guess.
well Manel,
i dont know what is going on, but i dont like it!
in response to Intel's email from my complaint about multi not working properly with Q6700 here i replied back as gtj suggested that the board was not working as expected or as advertised in their BIOS update notes, i got nothing.
after several weeks, and still no response, i sent another reply. see attachment.
i'm not sure what their problem is, but they need to pull their heads out of their...and start to be responsive to their customers.
sean
Sean
These guys at Intel all read from the same book, the replies are always the same. Funny how Intel states on their page that this is an extreme/enthusiast board with oc'ing capability but whenever you need tech support the first thing they say is that oc'ing is not supported. They should make up their minds on how the board is positioned. I guess oc'ing possibility is good for the marketing people since Intel is moving into the gamer/enthusiast crowd but when it comes to support these guys are not to be bothered.
I've been using Asus boards, no doubt they're good, but I got sick of all the flashy marketing buls**t, fancy technology names, weird color schemes, heatpipes, all the crap coming from the Taiwanese makers. This board seemed solid, the best looking I've seen, no legacy ports. Well, it's almost perfect. If only they used all solid caps, loose the Marvell controller and use the PCI Express lanes for the TI FireWire controller instead, better hardware monitoring and UEFI or include some missing BIOS options. I had issues with the onboard audio ports placement as well, my 900 audio cables are a little short to get there (Thanks Don at Antec for the new cables, he passed it on to engineering - that's service, behold Intel).
any real OC results on this board? any bench or any kind of review's? I even see it on ebay - but not a single review...
thx
one of the owners in this thread hit 500 FSB on this board, which is the max. i never tried to go that high because my CPU has a multi of 9 and i have no interest in theoretical limits. the highest i ever tried was 467 FSB for an even 4.2 GHz on my CPU and i hit it easily without having to raise any volts to an uncomfortable level. other than that, it's all pretty standard fare. it has eight USB ports in back and two eSATA, which is the kind of thing that's usually reserved for more expensive boards. my biggest gripe is the lack of solid caps and no automatically resetting BIOS when you set a bad OC and fail to post. you have to get old school and move the jumper, which is a royal pain imo.
want to take my q6600 to 333 fsb.
outside of disabling the eist and moving the fsb to 333. do i need to also take the memory to 1333? memory wise i have 4x1gb on the board, so i imagine that may also limit my over clock from what i have seen from other posts on this board and other mobos.
should i also leave the cpu voltage at default?
any suggestions or plan of action is greatly appreciated.
labrat,
you most likely will be fine without any adjustments other than a little vcore (at least in my case, had to bump a bit to get over 320, but maybe i have a bunk processor...seeing that core 3 likes to crap out in orthos anywhere over 362 when run with any other core or cores, but others fine...even all other 3 at once.)
can you give us some system specs?
bios revision
memory brand/model
cpu cooler?
PSU
etc.
bios 1521
mem team xtreme ddr3 1333 4x1gb TXD32048M1333HC7DC rated 777-21 2t. tho runnign it at 999-21 1t and 1.78v stable at the moment.
q6600
1000w silverstone psu
8800gt xfx
raid 0
thermaltake v1 cpu cooler
Anyone know if the X3360's are a go on this board ?
They should be fine.
you should be fine with everything you have with the possible exception of a bit more volts on the cpu. and as you have probably read previously in this thread, this board requires some patience while overclocking. for starters, make sure that watchdog is disabled in the maintenance BIOS. also, if you get no POSTs following clock increases, hard reboot a couple times, if still a problem, shut power of at PSU or unplug for a minute then power back up and try again.Quote:
Originally Posted by labrat
as i had mentioned before, with my q6700, which is essentially the same chip, i had issues around 320mhz and needed to bump the vcore some to get past it, so keep that in mind as you go, although it shouldnt be an issue until more like 350 (and it sounds like you dont plan on that) for you with the lower multi.
Well, just for the Heck of it I purchased the intel x48 MB. I'm not sure what drove me to it (probably boredom) but never the less its on the way. It should be here tomorrow from newegg.
It'll definitely cure your boredom. :)Quote:
Originally Posted by rchelicopter
We look forward to some side-by-side type reviews!Quote:
Originally Posted by rchelicopter
-bZj
Bios upgrade 1554 posted today.....many fixes
ooh, this is a big one. we now have C1E. multipliers fixed. underclocked CPU's don't have their multipliers change automatically. numerous IDCC issues fixed. and on and on and on. max FSB raised from 500 to 720!!! my god, it looks like they turned everything into sunshine with a single BIOS update. thanks, Sarge!
Big FIX C1E can now be disabled!
Did not fix boot problem where system will not cold boot but must
be shut down once and then will boot normally......
it fixed my soft reboot problem. no more blue screens when i restart from Windows. the release notes made it sound like they fixed the problem where it gets stuck in a power cycling loop when changing an overclock sometimes. i think this board went from zero to hero overnight. it looks like every problem i've seen mentioned in this thread has been addressed.
Wow, lots of stuff fixed. I'll be testing it on both the BT and BT2 this afternoon.
Perfect, my dx48bt2 is sitting on my doorstep at home waiting for me, I'll be trying that new bios as well..
They actually did something RIGHT! They moved the watchdog from the maintenance page to the performance page and they put C1E on the processor page.
And IDCC has also been updated to 4.1.0.168 beta (still), only works with BIOS 1554.
I almost thought Intel believed that the board with BIOS 1521 was already perfect, no new updates would be needed. Well, it seems they had been hard at work actually.
It seems they're reading this thread since they do address lots of issues reported here. We do send them our complains also, so not sure.
I was specially surprised with the PSB freq up to 720MHz, wow, do they believe this board can reach that high? That would be cool.
I'm not sure reagrding the tRD mentioned, this is usually the Performance Level, the PSB to mem MCH latency, that would be (hopefully) as low as possible but around 6-7 cycles, but hey set it to 0? Guess it's not the same thing.
Since my machine is stable without C1E disabled, should I just leave it on ?
i'm leaving mine on. i've grown to like it in the past few months. it hasn't affected my stability at all. i don't think it really helps temps either, but it doesn't hurt.
The only time I turn C1E off is when I'm running benchmarks and tuning. Otherwise, it's on. If you leave the vCore at stock, C1E will lower both the multipler and voltage when the cpu is idle. If you have vCore at soemthing other than stock, C1E will only lower the multiplier.
I just got my DX48BT2 up and running and everest still says its an X38 is this to be expected..
Here's a picture of my new X48 replacement, I decided to water cool the chipset this time..
Now that this board has been out awhile,
Can anyone comment on the throughput speed of the Marvell eSATA ports?
I am getting ready to buy an external HDD for back ups and am trying to decide between using the Marvell eSATA jacks - or - buying ae eSATA PCI bracket that runs a line into the internal ICH9 SATA headers.
Current Marvell config:
Enabled in standard IDE mode.
Using plain vanilla Vista / Microsoft PCI IDE drivers.
One DVD drive attached to the Marvell PATA connector.
the standard for eSATA is 3 Gb/s, just like the internal SATA connectors. i'm sure its throughput is a lot higher than whatever your hard drives can handle.
I need a little help here, Ive been trying to overclock with Bios's newer than 1478 and have been unsuccessful
Bios 1478 Settings
Fsb 450
Ref Freq 400
Mem Freq 1333
Mem Timings 7-7-7-20 2T
Bios 1521 & 1554 Will not operate at these settings, I think its the sub timings that are hanging me up but I have not been able to find the right settings.. It looks like with Bios 1478 my sub timings are 60,4,10,5,5 (from everest under chipset), I cannot run these timings with newer bios's and I've even tried 88,5,12,6,6 and still no go..
If anyone can suggest some different settings I would very much appreciate it, fornow I'm just sticking with ver 1478..