** OFFICIAL ** Intel DX38BT & DX48BT2 BoneTrail Thread

[Sean-E-Boy]

Updated LAN drivers download issue fixed

[Sean-E-Boy]

Correspondence with Intel

Subject: Updated IDCC for X38/48?


Any chance that there is a new IDCC coming forX38/48, or are you just going to drop us like a hot turd since X58 is out now? Would be great to control my system fans again!


From: idccfeedback@intel.com

Hi,

There is still a chance that we'll release a new version of IDCC for X38/X48, but it probably won't arrive before March. We had to make significant changes to IDCC to support X58 and some of those changes are not backward compatible with X38/X48. The X58 work should be finished by the end of January and then we'll circle back and try to make the current IDCC codebase backward compatible with X38/X48/G45. I know it's been a long wait, but I'm still hopeful that we can get you a fully functioning IDCC for your motherboard.

Thanks for your patience!

--MATT

Matt Alford | Software Product Manager| Intel Corporation

[Manel]

dl'ed v13.5 as well, will install now.
Regarding IDCC, well that's at least a reply with intention. However, reading this I feel that Matt here is screwing us around. I mean, "There's still a chance..."? Come on, it was supposed to exit at least a working final version of IDCC, or any software for that matter, to work with any product, not just beta. "... but I'm still hopeful tat we can get you a fully functioning IDCC for your motherboard.", man that almost makes me laugh! It seems that're making us a favour for having it working, maybe.
Come on Intel, you can do better than that.
You created high hopes for most of us here with this Extreme series of mobos and now let us down like this, great disappointment for many here. I guess non-Intel boards for most again from now on.
Also, seems that no new BIOS in the horizon as well, right? At least fix the power off issue, that's kind of embarrasing for you even.

[Sean-E-Boy]

It's amazing how this thread has just died in the last month.

Anyway, just thought I would share some GTL tuning experiences.

The first 2 GTL settings affect the Data Strobe (ref 0/2) and the Address Strobe (ref 1/3), respectively, NOT the cores or pairs of cores themselves as I had thought. In fact, these 2 GTL settings have nothing to do with the cores directly, but everything to do with the FSB and how the FSB detects what the cores are telling it.


Here is an analogy: If we think about this relationship between the CPU, FSB, and the MCH as a movie scene in a busy downtown setting in which the actor is speaking quietly, the scene of the movie becomes the entire signal that the CPU is sending to your stereo (the FSB) in order to transmit the actor's voice (the actual signal of interest) to the viewer (the MCH...could also be another core in the processor, but will assume for simplicity's sake that this is not the case here.)

Now since the scene was pre-recorded, the ratio of the volume of the actor's voice to the background noise cannot be adjusted. We can, however, turn up the volume of the stereo (vFSB) to make what they are saying more evident, and this works well in most cases. But at one point in this scene, a car horn (background noise) honks at the same time as the actor says a word, making understanding the word impossible no matter how loud (vFSB) you turn up the stereo.

So, you go to your bass and treble adjustment knobs (GTL refs 0/2 and 1/3) and adjust the frequencies (this isn’t exactly what happens, but works well) so that the noise introduced by the car horn is reproduced less by your stereo in relationship to the actor's voice and now the word can be understood by the viewers (the MCH).

But one of these viewers (the MCH) is a senior citizen whose hearing is going and sometimes cannot understand the words (doesn’t understand the signal sent by the FSB and therefore cannot transmit it to the ICH, RAM, etc.). But luckily, he wears a hearing aide (an amplifier-vMCH; and a tone adjustment knob GTL MCH REF) to assist him, but the components of the hearing aide are of lesser quality than those of the stereo (FSB) and provide only coarse tuning, but are effective.


Now back to computers:
As the speed of the FSB increases, so does the amount of background noise around the signal, now we can increase the vFSB to make this signal louder, or we can tune the GTL refs to better detect the signal, OR we can do both.

At "Auto" GTL settings, my Q9550 at 6x maxed stable FSB out at 472 with vFSB and vMCH maxed out, 2 dimms of ram fairly loose and @ 1:1.

Started tweaking the first 2 GTL REFs, gained stability, increased FSB, lost stability, repeated...now I hit a wall around 490, but it was different from the points of instability that were corrected or induced by changing the first 2 GTL settings. Those experiences resulted in windows freezing up at various points during loading (the longer it takes to freeze the better and adjusting up or down from the optimum will shorten this time), but this time the computer just failed to do anything other than reset or freeze at black screen once all of the POSTs completed. So I tried my hand at the GTL MCH REF setting and now FSB stable at 500 MHz @ 6x, but can’t get it any higher.

Started bumping up the multi as well as the vCORE. Not enough vCORE to remain stable at 500x8 with current cooling (liquid cooling on the way!) and wont run long at 8.5 multi. So I backed off the fsb some for now and running “stable” at 490x8 with vCORE 1.4375 in BIOS :: 1.41 in Everest @ idle :: 1.38 in Everest under load but still gets really hot, so no super long stress testing yet.
Will update next week once liquid CPU and NB cooling are installed.

[rchelicopter]

I hope i dont get in any trouble for this. I need to sell my Motherboard, RAM, CPU, and Waterblock (All detailed in signiture). The setup is approx 7 months old and I would like to sell as a combo but I will piece it out if neccessary. If interested please email me at rchelicopter@hotmail.com, thanks.

[Manel]

Sean, nice analogy! Keep tweaking and let us know.
I was also thinking that people here started to let go of the thread, too bad, too few posts. What's up?!

First drivers of 2009 out, Marvell RAID and Intel Matrix Storage.
No new BIOS though.

[Manel]

New nVidia 181.20 WHQL drivers out!
That was quick this time. For GTX285 and GTX295 support mainly I guess.
Some improvements in games, we shall see.
Still, PhysX is the older version unless it was just copy+paste from older drivers description and they forgot to correct.

[Mahyar.FF]

Sean, I liked your analogy.
Thanks

[Shemp]

Hi Sean-E-Boy & Petr0id....

Problem solved, the issue is this: I was using Windows XP Media Center, OEM build. The installation instructions with this are wrong. During the OS install, when the second disk is in, the instruction is to insert the SP2 disk, which was not included. I found a website, newegg discussion boards, which says to ignore what the instruction says, and to instead put the first CD back in the PC. With this, the OS loaded properly.


But boy oh boy I wonder if there is an icon for "what a pain in the you know what!"

The machine is happily running now!
thanks,
Matt

[Manel]

Intel download page down for DX38BT and DX48BT2, I wonder if there will be new things coming up... new BIOS maybe?!

[petr0id]

Nope! Still waiting for POST beeps fix...

[Manel]

Yeah, seems like it's still a no go.

[Sean-E-Boy]

Ok, now that my water cooling is installed I pushed to board pretty hard and this is what I got at 8.5 multi and a TON of volts (vcore in BIOS 1.5):

Can take the FSB much higher at 8X, but cannot reach this frequency.

So, with tuning the GTLs (most important one seemed to be the MCH) I was able to push my stable FSB @ 8.5x more than 15 mhz higher, and at lower multis more than 25 mhz higher as well as reduce the mems CAS to 6 from 7 WHILE at a higher frequency and still at stock voltages for the ram. Previously, a CAS of 6 would result in the system returning to default values in the OS for EVERY setting.

Havent started working the vFSB or vMCH down from max values...results soon.

Side note: seems that the Quads really stress the MCH, and however intel has it is set up in the dx38/48s is inferrior to how other mobo manufacturers set it up. Trolling other forums related to the x38/q9550 demonstrates that FSBs around 510-520 with the Q9550 of the same revision are not uncommon. Sure, it could just be my Q9550, but with the Q6700 I had previously, findings were similar that other mobos could take the FSB higher than intel's.

Sean-E-Boy

[Manel]

Marvell RAID again and imagine... Intel updated the INF to include all chipsets v9.1.0.1012, that was something!!! Remember we discussed this some time ago, not knowing which chipset drivers/inf would work with X38 or X48.

[Sean-E-Boy]

So for those of you interested in tuning your GTLs, here is what I did:

1. Since the first 3 GTL refs relate explicitly to the vFSB and use that value in calculating what the adjustments made in GTL refs mean, increase the vFSB to the maximum level that YOU are comfortable with.
2. Since increasing the vCore can make up for instability by misadjusted GTLs, drop it to stock (this is one source of instability that we want to keep or exaggerate, since the goals behind GTL tuning are to either make the processor operate at a given frequency with less voltage, or increase the stability of the FSB allowing it to be increased with less or equal vFSB).
3. Eliminate other sources of instability
a. drop the mem to 400x800 and loosen it (to 10-10-10-28-2T for example)
b. keep/raise the vMCH to a level that you know will be stable or that you are comfortable with...using the stock NB cooler 1.475-1.55, have a fan on it, up to 1.6 or so, liquid cooled, even higher.
c. Reduce your CPUs multi to its lowest value
4. Increase the FSB to the point where windows starts to boot, but fails to load completely (this is the point where signals from the CPU to the FSB and MCH are starting to become corrupted).
5. Restart the system a few times to get an idea of how long it takes from when windows starts to load until it freezes, restarts, BSODs. This should be pretty consistent and a good method for keeping track of the time is to watch the windows loading bar and count how many times it passes as well as how far into the bar it is when it freezes:

6. Once you have an idea as to how long it takes to freeze, enter the BIOS>Performance>CPU>Reference Voltage Overrides and increase the GTL REF 0 (this is the "Data Strobe") by 1. Save and Exit.
7. Let windows load up again and keep track of when it freezes. If you manage to make it into windows, restart and increase the FSB so that you dont...it is much easier to tune GTLs under a highly unstable environment rather than one that makes it into windows.
8. If windows loaded further than with the GTL REF 0 set where it was prior to step 6, increase it by 1 again.
9 Repeat 7 & 8 until windows load progress degrades, keeping track of the GTL setting that resulted in the "best" load progress.
10. Also try decreasing the GTL setting from where it was prior to step 6 and compare to increasing (for me, all gains were seen from increases, but this may not always be the case. After water cooling was added, GTL REF 1 and MCH GTL REF decreased by 1 or 2).
11. If 2 or more settings result in identical load progress, either pick the one in the middle (ie. if 10,11,12 were identical, choose 11) or increase the fsb by 1 or 2 to enhance the instability, possibly making these settings result in different load progresses.
12. Once you have found the optimal setting, move on to GTL REF 1, and then MCH GTL REF in the same fashion. MCH DDR REF in my case seemed to have little effect, but possibly if you are running a dual core and therefore stressing the MCH less from the processor you may gain a few Mhz of stability with high Mem Frequencies (ie at 1:2), but adjust this once everything else is adjusted.
13. Now from here you have to decide on what your goal is...a) high FSB and mem freq. or b) High processor freq. and there are many ways to go about this.
14. In either case, the next step would be to set your vCore to the maximum that you are comfortable with, but for:
a
1) keep your multi at its lowest and start increasing the memory frequency--slowest to fastest 400x800, 333x800, 400x1066, 266x800 (probably wont boot), 333x1066 (may not either), 400x1333 (booted in my case), 400x1600 (1:2), 333x1333 (1:2), 266x1066 (1:2).
2) if happy with mem frequency that booted, try increasing FSB again, but this time you are going to check stability within windows through prime, orthos, or with my favorite, the Intel Burn test (which will cripple your CPU within seconds or minutes, rather than minutes or hours as with prime or orthos), found here http://www.ultimate-filez.com/files/IntelBurnTest.zip. once you found max stable, start increasing your multi and stress again. If fine, increase multi again, if not, either decrease FSB until stable and then increase multi or see if MCH DDR REF makes a difference.
3) if not happy with the resulting mem frequency, reduce FSB and try to boot into next higher.
4) Once stable decrease timings (if you increased vDimm and changed MCH DDR REF, be aware that the MCH DDR REF uses vDimm as its reference, not vFSB and therefore changing vDimm may 'untune' MCH DDR REF) test again, repeat
5) Now stable with tight mem timings, begin to reduce vMCH and vCore, but do not adjust vFSB since GTL REFs you just found are dependent on the vFSB experience is showing that I may be wrong here...will update as I learn more

b
1) Increase your multi to the max (for non ES or Extreme chips) and try to boot, if no boot decrease FSB and try again (also keep in mind that it is possible that your highest CPU frequency could be found on the multi 1 step below the CPUs max multi)
2) If booted and stable in b1, increase FSB and stress again.
3) Once max stable found, start increasing memory frequency as in 'a1'
4) Decrease memory timing, then vMCH and then vCore. Do not, however, adjust vFSB since GTL REFs you just found are dependent on the vFSB experience is showing that I may be wrong here...will update as I learn more

Done

[fredoariaudo]

I like to buy 8 Gb of memory to my machine, I have:

Intel E8500
MB Intel X48BT2
2 x 2 Gb Supertalent DDR3 1333 Mhz
2 x MSI 4870 OC Edition
Seagte Barracudad 7200.11 500 Gb

Im planning sell my memory and to buy 4 modules of 2 Gb. What memory you recommend me to buy? 1333 Mhz or 1600?? Does this MB still have issues with 8GB DDR3 1600??

Thanks

[Manel]

Welcome Fred
You already have 4GB why not get another pair of 2GB of STs? Anyway, not a lot of people have 8GB, overkill for most. Does your 8500 get to 400FSB? If so 1600 could be the way to go if the board can handle it, it's a lot of load to the MCH.

[petr0id]

Why to run OC'ed machine, my friends? Is it just for fun, or it is because of something else?
My one is running great at default settings and i don't wanna more for now. DX38BT is a very good mobo. It works.

[Sean-E-Boy]

Same reason as why one puts a supercharger on their car...because you can...because there is more power to be had...and nothing makes you feel better than getting a lot more for very little.

sean

[redphil]

@Sean-E-Boy: Thanks for the guide.
I'll use it for pushing my q9300 a little bit more when I'll have time.

[HighTest]

Part of the reason may be that a number of previous DX38/DX48 users had decided to experiment with a DX58/Core i configuration? After all we are Xtreme Systems forum members and like to push the edge.

However, many others, such as myself, may have finally hit the sweet spot and have no further problems to report either. I have switched from a poster to more of a lurker simply as I've little more to contribute at this time.

I am encouraged that new driver updates are released, but I suspect it's more due to the fact that several of the chipsets have similar featuresets that are easy to create a common code base for. Matrix drivers for example or LAN drivers, these are common across a large number of the chipsets and are also used in the new DX58 chipsets.

If another BIOS update is released for our existing boards, I'll be very much encouraged! The BIOS has come a long, long way since the original purchase of my board. However, there are still some features that I and others would love to see, as we think this could help us wring even more OC goodness from these setups. Personally, how Intel continues with the support and continuing BIOS refinements of the DX38/DX48 motherboard owners will have a direct impact on a purchase decision of any future Intel motherboards.

The quality and layout of the motherboard was very encouraging to me, the hardware design team hit upon many of the qualities/features that I wanted. However the software engineers took that great design and fumbled a whole lot. It's a lot like the MB was a great Quarterback that saw a fellow team mate open and threw the ball to that player. Then it's like that player got distracted by flowering clover on the field or something and little if any yardage was accomplished. Sure there was some gain, but not anything like what the audience or the team expected beforehand.

Show me what is going to complete this with the existing DX38/DX48 to show that a great new gameplan is in play, before I shell out for the same old moves.

[Sean-E-Boy]

@ RedPhil: No problem. For my Quad, it seemed like the MCH REF was where all the performace gains came from, and if you are trying to keep your ram 1:2, the MCH DDR REF can help a little.

And for not changing the vFSB after tuning the REFs based on it I think I was wrong...have started reducing my vFSB since I had it at 1.5, still stable at 1.4 and will try reducing more and keep everyone posted as to if I had to change REFs.

I know that the purpose of the REFs is to make the signals readable, and that frequency affects readability the most, but not sure how lowering the vFSB will affect things since most of the REFs are based on the vFSB.

So your Q9300 has a 7.5 max multi? What FSB have you made it to at the highest and lowest multi with Auto REFs? At what voltages? x38 or x48? What cooler are you using?


@ HighTest: I hear you...One would think that since Intel makes all of the chips in these boards that really make a difference to OCers, and one would think that they know what the hell is going on with them, that they would grant us a little more tweakability, like the "Performance level" setting for the RAM, or the ability to run the memory faster than the FSB...or separate the two PCI-e slots by .5" more so if you have 2 double wide cards the second doesnt block the fan on the first.

Sean

Oh, and is there a way to make xtremesystmes keep me logged in for more than 5 minutes at a time? I have typing a reply, clicking "Post" and then having to log in again.

[Sean-E-Boy]

@ fredoariaudo: dont know about issues with 8gb ram and this board, but your selection on speed would depend on what FSB you are planning on using as well as what memory settings (1:2, 3:5, etc). Let us know that and maybe we can help.

All i can say otherwise is: What are you doing that fills up the 4 gb that you already have?

[Manel]

Sean
sent an email to idcc support as well and the info provided was the same you got some time back.
I do hope they go back to IDCC for our boards, and the BIOS guys should be doing their job as well.
If there is no development from Intel to fix the BIOS issues and finish IDCC I guess I'll never buy Intel boards again. Too bad, I was enjoying this one, even with the less than stellar features.

[Manel]

Intel's been launching new procs latelly, now the E8700 at 3.5GHz, cool. Also, the Quads with lower TDP some time back.
Even though they'll be in for their first bad quarter in 21/22 years (87 quarters I believe).