** OFFICIAL ** Intel DX38BT & DX48BT2 BoneTrail Thread

[Manel]

Nick
Been away for a while, reading just now.
My board is the DX38BT, running on Vista 32 SP1 and with the SMBus v2.0 drivers. I see most of the screenshots are only using the "regular" SMBus drivers that comes with the INF package. The v2.0 driver for SMBus I believe was installed with IDCC, not sure though. I have it installed and works fine, no error like the one you're reporting.
General->Device Status states This device is working properly.
Driver says Intel 1/1/2007 version 6.4.0.1 and Driver Details confirm this info.

Hope this helps.
Try installing the SMBus driver 2.0 and see if it works.

[Manel]

Seeing audio issues coming up, how do you guys find the audio drivers? It's not that they're not working for me, they are, but I don't get the Dolby tab in the IDT drivers, although everything seems to be in order. Does it only show if you have a DD receiver connected?
Also, the looks of the audio control panel is awful, those guys at IDT need a few lessons on GUI design I guess. I have seen info, long time already, on a v2.0 GUI, as opposed to the current v1.8, with newer drivers but so far haven't seen anything new. Also, the latest driver update had nothing new from the previous version.

[nick_DX48BT2]

Nick
The v2.0 driver for SMBus I believe was installed with IDCC, not sure though. I have it installed and works fine, no error like the one you're reporting.
General->Device Status states This device is working properly.
Driver says Intel 1/1/2007 version 6.4.0.1 and Driver Details confirm this info.

Hope this helps.
Try installing the SMBus driver 2.0 and see if it works.

Thanks a lot for the tip! I did see other people posting here with the 6.x version. However, the most recent one available on one Intel page (not yet on the DX48BT2-specific page) is version 8.6.x dated from 2008, which is more recent? That's the version Intel told me to use, not the one from the CD or on the DX48BT2 page, they said they had "some issues with it".

So the version I have should be the most recent version, based both on incremental numbers and revision date...? I mean, it would be in a perfect logical world

I did install IDCC previously on my previous Vista installation last week. I did not check SMBus after that, unfortunately...didn't know it came with drivers. I would rather avoid installing it, as others have stated here, there are some bugs with its behaviour after a while (beta version still), you shouldn't empty your user/local/temp folder, and there are some bugs in the Event log sometimes...plus it adds one more thing on startup

How did you find your drivers were revision "2.0", or that there was a revision "2.0" available? Did Intel say what issues were addressed with this new revision? I wished the Intel tech would have mentioned something about that or it could be on there website! (maybe I missed it?)

* Speaking of event log, I get a TCP/IP error "TCP/IP has reached the security limit imposed on the number of concurrent TCP connect attempts." It shows after startup. Even just after Vista installation and motherboard drivers. I don't have any virus (fresh HDD, scanned anyway), and no automatic program connecting to Internet really. Strange.

[Manel]

Hey Nick

I know some people have issues with IDCC, but it works ok for me. It's still beta alright, and it seems will be forever, but so far looks good. The guys at Intel must clean the house alright, every new version is better but there's always something... even the install paths keep changing. Maybe with the new BIOS revision coming up IDCC gets it's final status.
Regarding the INF (or chipset instalation utility as it's now called if I recall correctly) files, that's another Intel mistery. So many different versions and no one seems to know which is the latest or the right version for your mobo. I have also been around searching for the latest for 3 series (in my case) and this is different from the one on the CD as well as on their web page. Anyway, just decided to leave it as it is. The one issue have is that my SATA drives show up in IDE channels, not sure if this should be like that.
Regarding SMBus 2.0 driver, this what the description appears like in the System Devices in Device Manager after installing it "Intel (R) 82801 GB SMBus controller - 2930 (Intel (R) SMBus 2.0 driver). Reading the release notes I get even more confused, since the dates report back to January 9, 2007 and May 21, 2007. It states ICH9 support, ClearBus hang bug fix, WHQL certification, soft reset for ICH8 and ICH9. Go figure!!!
TCP/IP error? Could it be Vista connecting home, M$ is watching you!! Just kidding, not sure what could be, should check mine maybe.

[Randy]

Just a quick note here for people with S3 issues running Vista. S3 only works for me if the plan I choose is default. If I modify that plan - say to change standby from 1 hour to two, it won't work. Just set up a Windows Home Server in the past few weeks, and if that box is on, same thing, no S3 - so it may be a Vista issue. Just an FYI.

[Randy]

i have dual 4870's on my BT and i love them. i actually feel true love for these cards. CrossFire performance is awesome in games that scale with CrossFire. there are a few games that don't play well with it - those are the games i disable CrossFire for before playing.

download Fraps and run it in the background while you test games with CF enabled and disabled. the framerate counter will let you know if/how well CF is working for a particular game. also, make sure you're running the 8.6 Hotfix driver or 8.7 beta driver if you're feeling froggy. don't run the regular 8.6 driver with these cards. 8.7 will be the first official driver for the new Radeons.

Running the the Beta's and I am very pleased with the performance now over the 280GTX. I had to wait like a vulture on Newegg to get these, but of course, they are now readily available

[Manel]

Been looking again on Intel page, funny how they have the 9.0.0.1008 INF for every chipset but the X38 in the release notes, including the G31/33/35 and P35 as well as 4 series, but when you are about to download it the 3 series is present. These guys at Intel that update drivers in their page really don't give a s**t about us and try to keep things simple so as to avoid these types of problems.

[nick_DX48BT2]

Manel,

Thanks again for the tip with IDCC! The driver version with it seems older, but sounds like it could fix the bug I found. Funny Intel doesn't even know about it. There's more info here than at Intel!

I agree with you, the driver version mess from Intel should be cleaned up...even their tech people recommend different versions from different pages when you call

I researched more my TCP/IP error but it shouldn't apply in my case. I wonder how many people here have the same event. It appears in Event log sometimes after startup.

Liek you I also have my drive tagged as "ATA Device" in Disk Drives of the device manager, and in IDE ATA controllers, the first ATA Channel 0 shows that UDMA mode 6 is used under the Device box (doesn't show which drives) -- the other chanel 0 devices show nothing there.
I guess it's just how it's labelled. Nothing shows anything with the words "SATA" in my device manager actually.

[clark_carty]

Just a quick note here for people with S3 issues running Vista. S3 only works for me if the plan I choose is default. If I modify that plan - say to change standby from 1 hour to two, it won't work. Just set up a Windows Home Server in the past few weeks, and if that box is on, same thing, no S3 - so it may be a Vista issue. Just an FYI.

You have to go in the device manager, under the Intel Ethernet driver, and uncheck all of the wake on Magic Packet settings or else the WHS will constantly cause the machine to come out of S3. With everything unchecked, and PP1 installed on the WHS, it wakes it up just fine to do the backup.

[mryerse]

Speaking of event log, I get a TCP/IP error "TCP/IP has reached the security limit imposed on the number of concurrent TCP connect attempts." It shows after startup. Even just after Vista installation and motherboard drivers. I don't have any virus (fresh HDD, scanned anyway), and no automatic program connecting to Internet really. Strange.

That happens to me as a result of torrenting. Windows Vista (and XP SP2 or later for that matter) limit the number of TCP connections at a given time. The goal is to prevent worms from spreading really fast on the Internet. But a lot of people are upset because it makes their applications, such as torrenting, not work as well. And you can imagine Microsoft's response to torrent users who download legal content. They see torrenting as bad no matter what. The limit is set in a .sys file (tcpip.sys I think) which is binary so you can't just edit it. There are modified tcpip.sys files available, but I've tested them and had other problems as a result. So if you are torrenting, I recommend running Vista Ultimate over other Vista versions since it allows more TCP connections and just putting up with the limit.

If you are not torrenting, then I'd run some tests to see what kind of TCP connections you have, like netstat, or download tcpview from sysinternals.

EDIT: Okay, now that I've fully read your post I see it happens before you've installed anything. That is strange. It sounds like either an OS issue, or maybe a NIC issue, or mayve there is some app that's trying to connect. Again, I'd run some TCP diagnostic tests like tcpview or something.

[nick_DX48BT2]

EDIT: Okay, now that I've fully read your post I see it happens before you've installed anything. That is strange. It sounds like either an OS issue, or maybe a NIC issue, or mayve there is some app that's trying to connect. Again, I'd run some TCP diagnostic tests like tcpview or something.

Yes, indeed, it happens before any program is even installed. Plus I have Ultimate...I tried to run some diagnostics and track which IP addresses were used but it wasn't conclusive, so I'll have to dig deeper. I wondered if anyone had seen the same message before even installing programs...

[gtj]

Yes, indeed, it happens before any program is even installed. Plus I have Ultimate...I tried to run some diagnostics and track which IP addresses were used but it wasn't conclusive, so I'll have to dig deeper. I wondered if anyone had seen the same message before even installing programs...

Do you live near a nuclear power plant or over an MRI scanner or something?? Strange things seem to follow you around.

From a command window, type 'netstat -n' and you'll see a list of all your open connections along with the remote ip address and ports.

[nick_DX48BT2]

Strange things seem to follow you around.

I know!!!! I've built many many computers, but never had anything "strange" like the few issues I've had here...just the normal Windows bugs and hardware limitation or failures

Well, at least I've determined that the motherboard POST bug fater applying power, and the SMBus driver bug & Vista message I found are not limited to me, it's just a plain old bug.

The last on my list was the TCP/IP error in the event log... I wasn't too concerned about it, but on the other hand it does not seem it should be there before any program is installed after Vista installation. I've tried netstat before, and tcpview yesterday based on the suggestion from mryerse...there are a few processes, not all easy to identify, and the TCPIP error occurs after startup but it can be 2 min or 2 hours...so I have to be patient, monitor all and hopefully I'll see something to be able to understand.
I also found a lot of Time_Wait, from SystemProcess [0]. For example if I launch Firefox, start browsing, I see connections to random servers, some I am browsing some not but I assume it's connections for things like links, ads, etc on the page I am viewing. Then when I close, the firefox processes close, but those other stay there as Time_Wait (for example Google.com, some IP addresses, etc), but they are under SystemProcess [0] in tcpview. Maybe I'll use netstat and the PID to see better which process is being used.

So I wonder if others see that too...I mean, this error message but without any program installed like torrents...it's easy, you can go back to the first few events after Windows installation, you'll see if it's there already, that's how I found it was there early on ;o)

[gtj]

Well, I don't see the error message but the TIME_WAIT entries are normal. Windows defaults to a 240 second delay after a connection is closed before it flushes its resources. Google "TcpTimedWaitDelay". There's a registry key you can set to reduce the delay. I normally set mine to 30 seconds.

[nick_DX48BT2]

Well, I don't see the error message but the TIME_WAIT entries are normal. Windows defaults to a 240 second delay after a connection is closed before it flushes its resources. Google "TcpTimedWaitDelay". There's a registry key you can set to reduce the delay. I normally set mine to 30 seconds.

Thanks! I did found that yesterday after a quick search, but wasn't sure if getting a few Time_Wait entries was normal in the first place...

So I'm really the only one with the TCP/IP error??
Could it be my network/ISP, router too? I'll have to try with another PC, either Vista or XP SP2 or 3 (SP1 doesn't have the limit if I understand correctly).

PS: my old PC (XP SP1) is still working fine without any bugs. I miss it ;o)

[Loony]

How does this board clock e8400's under water?

Also, how does this board clock DDR3? And which DDR3 kit works the best (preferably 2x2gb kits.)

[xMrBunglex]

How does this board clock e8400's under water?

Also, how does this board clock DDR3? And which DDR3 kit works the best (preferably 2x2gb kits.)

fine and fine. your E8400 will hit its wall before the motherboard hits its FSB wall. i've known people that ran their memory as high as DDR 2000 on this board. there is no real reason to go higher than that unless you're just looking for bragging rights.

[down8]

It doesn't like my 2x1GB Ballistix 1600 kit. It won't even get to 1600. I have to run it at 1333.

It wouldn't run 2x1GB Aeneon XTune 1333 at all. Just refused to POST.

-bZj

[Loony]

fine and fine. your E8400 will hit its wall before the motherboard hits its FSB wall. i've known people that ran their memory as high as DDR 2000 on this board. there is no real reason to go higher than that unless you're just looking for bragging rights.

My goals were 4.5ghz and DDR3 2000 so hopefully I'll be able to achieve that!

I was thinking about using a 2x2gb kit from either G.Skill (I have a lot of good experiences from them) or OCZ.

EDIT: I'm guessing you were referring to the X38 version, am I correct?

[gtj]

My goals were 4.5ghz and DDR3 2000 so hopefully I'll be able to achieve that!

I was thinking about using a 2x2gb kit from either G.Skill (I have a lot of good experiences from them) or OCZ.

EDIT: I'm guessing you were referring to the X38 version, am I correct?

500x9 might be pushing it a little, at least with the current BIOS and I don't think you'll get that on the X38. With the X48 and an upcoming BIOS, should be do-able. Tony from OCZ has been working with Intel on the BIOS and has reported some encouraging test results.

[Avantlord]

hi! i have this mobo too, the x38 one, and i´m searching some good DDR3 1600, i was thinking in OCZ Flex CAS 6 and the Gskill cas 7 ones.... does anyone can help me XD

[nemesis326]

I installed my DX48BT2 and Windows Vista 64bit service pack 1 yesterday, immediately I saw in the device manager that the SMBus controller had a yellow exclamation mark.

So I went to the Intel website and downloaded the chipset driver version 8.3.1.1010 and after restarting Windows the device disappeared.

If I right click the SMBus controller and check the properties, it says "No driver installed", however, if that was the actual case, Windows should still be reporting the SMBus controller with a conflict (!) and it is not.
I haven't see any pop-up configuration wizard asking for drivers since...

I'm a little concern about it, but other than that, my system is running like a champ, I had to work with Kingston to clock my memory at 1600mhz since later on I discovered that most of nowadays memory by factory defaults kicks at basic frecuencies to ensure POST.

Hope my experience helps you clarify something with the issue reported by nick DX48BT2...

[gtj]

hi! i have this mobo too, the x38 one, and i´m searching some good DDR3 1600, i was thinking in OCZ Flex CAS 6 and the Gskill cas 7 ones.... does anyone can help me XD

I've always liked OCZ. Tony from OCZ visits here occasionally and he's also working with Intel on the BIOS for this board.

I installed my DX48BT2 and Windows Vista 64bit service pack 1 yesterday, immediately I saw in the device manager that the SMBus controller had a yellow exclamation mark.

So I went to the Intel website and downloaded the chipset driver version 8.3.1.1010 and after restarting Windows the device disappeared.

If I right click the SMBus controller and check the properties, it says "No driver installed", however, if that was the actual case, Windows should still be reporting the SMBus controller with a conflict (!) and it is not.
I haven't see any pop-up configuration wizard asking for drivers since...

I'm a little concern about it, but other than that, my system is running like a champ, I had to work with Kingston to clock my memory at 1600mhz since later on I discovered that most of nowadays memory by factory defaults kicks at basic frecuencies to ensure POST.

Hope my experience helps you clarify something with the issue reported by nick DX48BT2...

That's normal. There actually isn't a driver. The chipset "drivers" are really just a collection if INF files that describe the hardware.

[Avantlord]

OK i´ll wait for tony´s opinion about the 1600Mhz flex cas 6, they are very good looking

[nick_DX48BT2]

Hope my experience helps you clarify something with the issue reported by nick DX48BT2...

Thank you so much for your post!!! Let me update you and the others on the situation:

1. First of all, thanks to you and the others for replying to my original post(s). This confirmed that this is a real bug, not limited to one particular case, mistake, or configuration.
2. This has been confirmed here on Windows Vista SP1 AND XP. Don't know about Vista w/out SP1.
3. This happens no matter what you do, no other devices installed, just Vista and the board drivers and no other drivers, and whether you use the 2007 drivers from the CD or the DX48BT2 web page (8.3.x series), or the new 2008 drivers series 8.6.x included only in the 9.x June 2008 package available only on one page of Intel's site dedicated to Intel Series 4 motherboard (X48). Some techs at Intel are aware of the page and ask to install only those drivers, some other techs don't know about it, and finally the DX48BT2 page has not yet been updated. It's a driver mess. Anyway, it doesn't change the bug.
4. Uninstalling reinstalling reformatting etc etc will not change the bug. Intel reports the installation successful, clean install or not, new or old drivers. I've done many tests.
5. This issue is not observed by people who have the older 6.x series driver from 01/2007, that apparently is installed with IDCC, but yet has the name "SMBus 2.0" (vs. "SMBus").
6. The popup message shown by Vista is random. It may or may not show on your system. It does not stay on long, you can miss it. It appears only once. It is normal to see it given the bug that the SMBus device says "no driver installed"!
7. This is without doubt an Intel bug. Either there is a driver installed, or there isn't. So the fact that no exclamation point are shown in the device manager, and driver date/name/version are provided, conflict with the fact that the status says no driver installed, and that clicking driver details show a message saying that no drivers were loaded.
8. Now, the real news: I've been on the phone, email, etc with Intel for more than a week. I've done all the tests they asked me to do, half of them were not relevant or smart suggestions, but I did it anyway. I asked them to escalate, because they were concluding by: sorry it's just you, or you could try to RMA your board.

[Ignore the message below -- see my updated post 7/18/2008 at 6:57PM, http://www.xtremesystems.org/Forums/...postcount=1314
The situation was handled better by Intel after I posted the information below, and I do think they read it (see my post 7/18/2008 6:57PM for the reason/details)
I've finally gotten a call back. Read on, you will be amazed:
a. A voice mail tells me that the error I found could not be duplicated on Vista SP1 on their systems, and asks me to call back. I have saved the voicemail and can post it if needed.
b. I talk to yet another Intel tech support guy, explain the situation. He says basically it's just me with the bug, no one else. In order to explain the situation better since he's not really top-notch up to date with the situation, I ask him to come to this forum to see the posts, and screenshots of the bug. OK, now he puts me on hold to go talk to other people, for 5-10 minutes.
c. Finally gets back to me. Funny, now the story has changed: they were in fact able to replicate my bug!? He tries to explain a few things that are completely incoherent, then goes on to say that if I want to get rid of the message, I should try to install a RAID system, it could work, because of different drivers. I explain that I do not want to buy another hard drive and RAID it to see if it solves their bug, but am asking simply if this bug is an issue or not. He does not answer the question, and immediately says: it's not an Intel bug, it's Microsoft, it's not our problem, you need to talk to Microsoft.
d. I explain calmly again that this is visible BOTH on Vista and XP, specifically with their drivers, the one that come with the board or on their website, and no other board on the market has this bug, so really, it's up to them to investigate. I tell him again that I simply want to know if I can safely ignore the bug, and if there will be an update, investigation, or maybe a note on the FAQ section of the Intel website.
e. Now he suddenly changes story again, and tells me that I'm really the only one with the bug, "no one else has reported it".
f. At this point, I am wondering how many times he is going to change stories, and if he is under the influence of drugs. So I again explain him that he just said they were able to replicate the bug, that it is posted here by others, and that the DX48BT2 has not been on the market for long (although same bug with DX38BT2). I know he is lying because I know personally someone else who has also wrote to Intel to mention the bug, with an assigned case number. I have his case number too.
g. No hope at this point, I ask if I can talk to someone else or if the case can be reviewed and escalated.

I tell him one more time that I have seen an error message in Vista, have seen the Device Manager bug, and other have seen the same thing, so I simply want to know if they have confirmed the bug and if I can simply ignore it or if they are going to work on it, that's it, I'm not asking for the moon! For God's sake, it's Intel most advanced "Extreme Series", most expensive, most recent board for a company that people buy because of engineering/technical support and reliability vs. some cheaper overseas boards! From the world leader in semiconductors, they make the board, the northbridge and southbridge, and the processor too!!! And their profits are up 25%, announced two days ago(http://money.cnn.com/2008/07/15/tech...ion=2008071518), it's not like they're going under.

I'm sad that the phone conversation was not recorded. I've called Intel before and it was handled in a much more professional way, no complaints. For this case, the previous guy that escalated the issue wasn't great either, and this last one was definitely unprofessional and/or really incomptetent. I will not give his name because it's likely not his real name -- please note that I don't think that the fact that English was not his mother tongue did not create any communication issues, there was no miscommunication. He was just trying to blame anything on anybody, "Intel has no responsibility".

So in the end he said that he would put a note to escalate the issue after I insisted, but "probably nothing would happen", and that "no, he can't say the bug is not an issue, it's not true, I shouldn't have the bug, he says there IS an issue" but tells me I'm on my own. Wonderful. Poll: Should I buy again Intel in the future? (no need to reply )

Note that I also sent an email/letter to Intel, but save your some time: they handle by email only simple requests. When it doesn't work or if there is a real bug, they will eventually ask you to call them.

PLEASE, IF YOU ARE READING:
- Call Intel at (916) 377-7000
- They will ask you first an email address where they can reach you, your motherboard, and give you a case number
- You will be transferred shortly to the Tech Support line; when you get the guy/girl on the phone, reference immediately this thread, specifically page 51 with the screenshot: this will save you a lot of time in explanations

Hopefully if a few people call they will at least do something.

Hopefully someone at Intel who handles customer satisfaction or engineering development or marketing of their top-of-the line boards will read this forum...nooo, who am I kidding You can find it with Google though...]