*** Official MSI 875P Neo-FIS2R/LSR (875P) Thread ***

[shawnjackson]

My MB is coming in on Wed... and I will be using 2 SATA WD Raptor drives on RAID 0... so my question is...

Do I use the Intel SATA controller or the Promise and why?

 

[Oddish]

I just thought I'd tell those who don't know that MSI just released a new BIOS (1.3) which supposedly fixes the overclocking, memory and temperature issues.

 

[kaps]

I just got my MSI 875P Neo-FIS2R and installed it on Saturday with a P4 3.0GHz 800 FSB chip and twin Corsair PC3200 CL2 DDR sticks. The system is very fast but I wanted to compare it with my old AMD 2400 / ASUS A7V333 system, so I ran Passmark over the new system to get a sense of my new speed. I was Very surpised to see that the difference was not that great. The old system scored 411.26 and the new system scored 466.55, a scant 13.44% increase. Below are the results (the AMD system first).

Comments welcome!

Karl

Benchmark results
Test Name NewBaseLine_02222003 This Computer
Math - Addition 490.24 734.92
Math - Subtraction 486.68 734.77
Math - Multiplication 430.19 557.96
Math - Division 44.63 51.84
Math - Floating Point Addition 436.94 636.41
Math - Floating Point Subtraction 436.23 637.14
Math - Floating Point Multiplication 437.26 619.90
Math - Floating Point Division 113.34 77.89
Math - Maximum MegaFLOPS 657.38 236.07
Graphics 2D - Lines 171.50 114.85
Graphics 2D - Rectangles 162.25 162.32
Graphics 2D - Shapes 34.52 26.42
Graphics 3D - Simple 1582.59 1694.40
Graphics 3D - Medium 1197.85 1283.75
Graphics 3D - Complex 615.19 653.97
Memory - Allocate Small Block 1349.50 1016.57
Memory - Read Cached 669.10 1549.53
Memory - Read Uncached 463.81 1424.15
Memory - Write 430.34 1104.62
Disk - Sequential Read 34.30 43.78
Disk - Sequential Write 24.63 34.23
Disk - Random Seek + RW 6.78 3.79
CD - Read 4.45 2.81
MMX - Addition 702.47 663.19
MMX - Subtraction 694.43 659.29
MMX - Multiplication 679.33 648.43
MMX - SSE/3DNow! 1246.49 1465.29
Math Mark 364.95 442.84
2D Graphics Mark 491.03 404.78
Memory Mark 283.99 496.75
Disk Mark 173.48 215.95
CD Mark 222.48 140.54
3D Graphics Mark 1131.88 1210.71
MMX Mark 422.81 437.26
PassMark Rating 411.26 466.55

System information- This Computer
CPU Manufacturer:GenuineIntel
Number of CPU: 2
CPU Type: Intel(R) Pentium(R) 4 CPU 3.00GHz
CPU1 Speed: 3000.2 MHz
CPU2 Speed: 3000.1 MHz
Cache size: 512KB
O/S: Windows XP
Total RAM: 1073197056 Bytes
Available RAM:696815616 Bytes
Video settings: 1024x768x32
Video driver:
DESCRIPTION:
ALL-IN-WONDER 9700 SERIES
MANUFACTURER:
ALL-IN-WONDER 9700 SERIES
BIOS:
BK-ATI VER008.004.008.000
DATE:
02/10/21
Drive Letter: C
Total Disk Space: 120.0 GBytes
Cluster Size: 4 KBytes
File system: NTFS



--------------------------------------------------------------------------------

Results Produced by; PassMark PerformanceTest (http://www.passmark.com)
Version: V4.0

 

[kaps]

I just looked at the Frequency/Voltage Control settings and while my DRAM is running at 400, my CPU Bus Clock is only at 200MHz. Is this correct?

Karl

 

[Jeriko]

Using the CoreCenter utility, what voltage settings would you all suggest for 3GHz, the ~3.1GHz range, and ~3.2GHz ranges? I'm not all that familiar with overclocking, and I think when I push it up to 3.0GHz my 2.8 isn't getting enough juice. Systems gets unstable.

Edit: Could it possibly be the RAM timings?

-J

 

[thermite88]

Originally posted by: o1die
This issue may have already been covered, but for those who missed it, the 865p and 875p chipsets will support 400 fsb northwood cpus (512 cache), but not the willamette cpus (256 cache).

I will call it limited support.

You can run the Northwood at 100MHz bus, but the CPU/MEM frequency ratio is fixed at 3:4. You cannot change it to 1:1. It means that you cannot overclock the CPU bus to much more than 120MHz before you hit the limit of the memory which will be running at 160MHz.

On the contrary, many A series Northwood run happily byond 160FSB in 845 boards using 1:1 CPU/MEM ratio.

 

[smahoney]

Quote:

http://www.theinquirer.net/?article=9572
<SPAN id=article_body_title>
MSI Canterwood BIOS kills mobo claim</SPAN>

<FONT size=4>Users up in arms, incandescent, and the like</FONT>

<FONT size=2>By<A id=article_body_authormail title="email the author" href="mailto:newsdesk@theinquirer.net"> <SPAN id=article_body_author>INQUIRER staff</SPAN></A>: <SPAN id=article_body_submitted>Tuesday 20 May 2003, 09:45</SPAN></FONT>
<DIV id=article_body_content><B>END USERS WITH</B> i875P (Canterwood) motherboards are claiming that the latest BIOS from Microstar International (MSI) makes their FIS2R-NEO products like Caesar. Dead.
Reactions on the MSI forum to the problem flashing the BIOS range from mild ire to intense incandescence.
One user said that version 1.3 of the BIOS was intended to fix some problems between mobo and 400MHz DDR memory and remove some high temperature readings.
But after apparently flashing the BIOS, some users find themselves in charge of what appears to be a hefty hunk of expensive real estate that's dead to the world.
There's some comments about this problem , while the link to the BIOS page can be found <a href="http://www.msi.com.tw/program/support/bios/bos/spt_bos_detail.php?UID=434&NAME=MS-6758" target=_blank>here.
We're attempting to get a response from MSI at press time. µ
A friend is looking at the board and I wanted to know if anyone here has experienced this issue...</DIV>

 

[kaps]

I flashed my BIOS yesterday with the new release and didn't have any issues at all. It worked perfectly. I did not notice and increase is system speed though.

Karl

 

[Jeriko]

I also flashed the BIOS - from 1.2 to 1.3 (I skipped the beta) and have no troubles so far. There was concern in the MSI tread that it might be related to fast memory timings - but I have Muskin Level II (2-2-2).

Maybe I'm just really lucky? Or maybe the problem only occurs whn flashing from the beta 1.3 to the final 1.3?

-J

 

[SleepyGreggy]

The 1.3 was taken down from MSI bios website??

 

[kaps]

Yes, I got mine from the MSI site.

Can anyone help me with my question on system bus speed. Mine reads 200 right now but my benchmarks don't make sense, so I think I have something wrong.

Thanks in advance!

Karl

 

[SleepyGreggy]

DO NOT FLASH WITH 1.3 BIOS, IT MIGHT KILL YOUR BOARD.
http://www.msi.com.tw/program/e_service/forum/thread.php?threadid=19744&boardid=10&styleid=1&sid=a1e2ed3abb150f2b25d0a28317142dbe

ALSO REPORTED ON THEINQUIRER.NET

 

[mac35]

Kaps,
The CPU Bus clock should be at 200, intel's front side busses are quad pumped so the 200x4 = 800mhz FSB.


Jeriko,
I would leave the voltage settings alone when you overclock unless your system is having stability issues. Thats just my opinion, however, when you get up to the 3.2 range etc you probably will have to bump the voltage up.



Wow I'm glad I decided to wait before flashing my BIOS to version 1.3. I was going to do that yesterday but decided to wait and read more feedback, glad I did. MSI has taken version 1.3 down until they resolve the problem so hopefully no other boards will be damaged.
Mac

 

[Jeriko]

^ I have stability issues over ~211 FSB - which isn't much of an overclock. My vcore setting in the Corecell utility is 1.5250 (seems to run at 1.49 most of the time though), and my memory voltage is 2.60.

-J

 

[SleepyGreggy]

This mobo looks promising but many issues seem to be plaguing it. I guess I will wait until the 13 motherboard roundup to get the board that suits me best.

 

[BeefE1]

I just built a brand new MSI 875 Neo-FIS2R wtih a P4 3.0Ghz 800Mhz FSB with 2 Kingston 512MB 400Mhz modules and have gotten problems running it with the DDR enabled. I can run it with both DIMMs in slot 1 and 2, but if I run them in slots 1 and 3 or 2 and 4 I get problems:
Examples:
BEFORE BIOS upgrade. The DDR worked, ..sort of. Everything worked fine until I tried to run UT2K3, then it ran the game for about 5 seconds, then it would freeze the whole machine. We could put the memory in slots 1 and 2 nd it ran fine.
We then went to MSI and updated the BIOS to version 1.3. The whole machine crapped out entirely after the BIOS update. This BIOS update was supposedto fix problems with some 400Mhz DDR memory modules. It did not. We had to rebuildthe who OS from the ground up (WinXP Professional). Once everythign was back up adn running with the ne BIOs, we could not even get teh machine to boot with the DIMMs in slots 1 and 3 or 2 and 4. bUt if I placed both DIMMs in slots 1 and 2, everythign worked perfectly.
Went back to MSI's site yestyerday and they have REMOVED the BIOS version 1.3, and only have 1.0, 1.1 and 1.2. Os I guess tehy got the same or simialr problems as we did.
So I guess we just sit and wait until a new BIOS updte comes out, or we will have to buy all new memory.

 

[kaps]

Memory timing is critical when running dual DDR. It is suggested that one buys matched sticks, where they have been tested and the timing is within very close tolerances.

Karl

 

[BigBlueHeaven]

A new BIOS update tooday 1.4 finaly

I haven't tried it yet but I will later today, and hopefully it will solve all my problems with my new bought MSI 875P NEO LSR

Link:
http://www.msi.com.tw/program/support/bios/bos/spt_bos_detail.php?UID=434&NAME=MS-6758

 

[ccgr]

MSI board caught over clocking!

http://www17.tomshardware.com/cpu/20030522/index.html

 

[Budman]

Originally posted by: ccgr
MSI board caught over clocking!

http://www17.tomshardware.com/cpu/20030522/index.html[/q

They are calling it a new feature.

Here's what MSI has to say.

"To the Editors of Tom' s Hardware,
We are writing in response to your article "Intel Rigs Up: P4 Series with FSB800"

Indeed what your lab engineer discovered is true.

MSI does plan to use the said "logic circuit" or "dynamic overclocking" as a feature for this board.

The reason why we haven't announced it to anybody yet is because we are in the process of doing internal testing and applying for a patent.

This feature should be available for high-end MSI motherboard in our next BIOS release.

MSI's "dynamic overclocking" (the feature name still has not been finalized as of this writing) feature is exactly what your article described."The manufacturer has incorporated an ingenious logic circuit that increases the FSB speed between 6 and 8 percent while programs are running; the processor is automatically overclocked. However, FSB and CPU speeds are only increased when applications are started or when benchmark programs have finished - subject to CPU usage reaching close to 100 percent. And this is really interesting: the overclocking does not show up using conventional benchmarking utilities such as WCPUID, Intel CPU Frequency Display, CPUZ or SiSoft Sandra 2003. You need a very specialized tool to reveal the increased bus speed."

The reason why the overclocking does not show up on well-known utilities such as WCPUID, Intel Frequency Display, CPUZ and SiSoft Sandra 2003 is because these programs cannot detect dynamic overclocking. That is why you need a special developer's tool to detect dynamic overclocking in realtime.

We are also planning to incorporate this feature on MSI boards in the future as well, and we will offer an option to activate and deactivate this feature on these boards.

Once again, thanks for everything!!!

Best Regards,
MSI Marketing"

taken from here 865pe auto overclocking?????

 

[Jeriko]

Well.... BIOS 1.4 killed my NEO-LSR.

This is ridiculous. 1.3 worked for me, while for a lot of others it disabled their board. 1.4 seems to work for most people, but it disabled mine. I shouldn't have to put up with killer BIOS, worrying if every time I flash will be my last. If I can't get this board up and running tomorrow, I'm taking it back for... Well I don't know what. Probably the Abit. It may not be the greatest board, but at least flashing the BIOS won't kill it.

-J

 

[kaps]

BIOS update 1.4 worked fine for me, as did 1.3.

Karl

 

[DamageInc]

What a pain in the ass this board has been for a lot of people. I am still hanging on to BIOS 1.0 for dear life, but if it is true that these "new and improved" BIOS revisions are killing people's boards, well then that's just horrible.

 

[Jeriko]

^ The problem is that the new BIOS are setting the memory voltage to 2.5 and it's just not getting enough juice. I finally got my Neo-LSR back up and running after an AMI BIOS recovery process, clearing the CMOS, and replacing the PC3200 RAM with an old stick of Crucial 2100 that could work with 2.5 volts. That got me into the BIOS long enough to adjust it to 2.6 - which got the PC3200 Mushkin back up and running, and the board functional again.

But god, I'm just not going through this hassle again. There are too many bugs on this board to be fearful of BIOS flashes - too much needs to be fixed. But if I have to worry about killing it with every flash, and going through long drawn out processes to get it working again, it's just not worth it.

Worse yet, not everyone has old low voltage-friendly sticks of 2100 laying around like I do, and recovery simply wasn't possible without it.

-J

 

[Jeriko]

By the way, let this be a lesson to everyone. ALWAYS HAVE A FLOPPY DRIVE. Even if you don't keep it plugged in and installed, keep it around for when you need it. I see a lot of folks questioning their need these days - but there would have been no other way for me to download the old BIOS and rename the file for the recovery process on my notebook, and then get the file to my new system, without one.

-J