Joshua7
asked on
Compaq Deskpro 4000 6300MMX DIP switch settings
The abovementioned computer has a DIP switch. A label inside the cover and the manual both list the following settings:
2 3 4 5
F F N N = 233
N N F N = 266
N F F N = 300
F N F N = 333
The machine came with a P2/300 slot 1 and a huge heatsink. I have installed a P2/350 with a heatsink/fan. The machine boots fine, and I have even flashed the BIOS to the last release with the new processor in place. The problem is that the processor is reported as either a 210 or a 233, depending on how the switches are set. This is true for all 16 combinations. My questions are:
1) Is this normal for a Deskpro ?
2) Is the processor not functioning at its rated speed ?
3) Is there a way to get the BIOS to tell the truth ?
4) Does anyone have an understanding of what those DIP switches really do ?
2 3 4 5
F F N N = 233
N N F N = 266
N F F N = 300
F N F N = 333
The machine came with a P2/300 slot 1 and a huge heatsink. I have installed a P2/350 with a heatsink/fan. The machine boots fine, and I have even flashed the BIOS to the last release with the new processor in place. The problem is that the processor is reported as either a 210 or a 233, depending on how the switches are set. This is true for all 16 combinations. My questions are:
1) Is this normal for a Deskpro ?
2) Is the processor not functioning at its rated speed ?
3) Is there a way to get the BIOS to tell the truth ?
4) Does anyone have an understanding of what those DIP switches really do ?
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
KronosTM - good to know that it will run better
tmj883 - there's an important piece of info. I was checking the speed reported by a program called Belarc Advisor (from belarc.com) which usually does a good job of telling you the truth, but really fell down here. When I checked the speed using the Compaq Setup utilities, here's what I found:
2345
FFFF -
FFFN -
FFNF 210
FFNN 233
FNFF -
FNFN 333
FNNF 180
FNNN 200
NFFF 270
NFFN 300
NFNF 150
NFNN 166
NNFF 240
NNFN 266
NNNF -
NNNN -
From what you said I gather that if I select one of the switch combinations that don't report a speed, but simply list the P2 processor, that the machine will slow to its lowest speed, which is actually 150. Do you know this for a fact ? Is there a good reporting tool out there which accurately measures the processor speed, regardless of what the BIOS might say ?
I was also hoping for a specific explanation of what each switch affects, but I guess that this is a deep dark Compaq proprietary secret, unlike other manufacturers who tell you which switches and jumpers set the speed, multiplier etc. The board has an Intel 440LX chipset, but is probably an OEM job, and won't be documented on Intel's website.
tmj883 - there's an important piece of info. I was checking the speed reported by a program called Belarc Advisor (from belarc.com) which usually does a good job of telling you the truth, but really fell down here. When I checked the speed using the Compaq Setup utilities, here's what I found:
2345
FFFF -
FFFN -
FFNF 210
FFNN 233
FNFF -
FNFN 333
FNNF 180
FNNN 200
NFFF 270
NFFN 300
NFNF 150
NFNN 166
NNFF 240
NNFN 266
NNNF -
NNNN -
From what you said I gather that if I select one of the switch combinations that don't report a speed, but simply list the P2 processor, that the machine will slow to its lowest speed, which is actually 150. Do you know this for a fact ? Is there a good reporting tool out there which accurately measures the processor speed, regardless of what the BIOS might say ?
I was also hoping for a specific explanation of what each switch affects, but I guess that this is a deep dark Compaq proprietary secret, unlike other manufacturers who tell you which switches and jumpers set the speed, multiplier etc. The board has an Intel 440LX chipset, but is probably an OEM job, and won't be documented on Intel's website.
This mainboard supports 60 and 66 mHz FSB speeds, this is why there are odd cpu speeds available like 180 mHz. The DIP switches controll both the FSB and cpu multiplier on this old mainboard. Most modern mainboards have separate, and/or automatic switches. The difficulty is that the 350 cpu uses a 100mHz FSB. The cpu most likely has a locked multiplier(3.5X100=350mHz) that cannot be changed. The only FSB settings available are 60 and 66 mHz...(3.5X60=210) and (3.5X66=233)...this results in the slower clock speeds than the cpu is rated for. So after all, Belarc did not fall down... Get the correct processor...66mHz FSB(according to Compaq, up to 366mHz but this is a 66mHz with a multiplier of 5.5)...if you want to go faster...hope this helps you to understand...
1.the mainboard supports only 60 and 66 mHz FSB speeds.
2.the cpu(new) is multiplier locked and requires a 100 mHz FSB to run at 350 mHz.(3.5X100)
3.Compaq provides extensive coverage of legacy equipment like this and specific descriptions of the switches are available. Intel can verify the PII 350 is 100mHz FSB.
Very different cpu...from PII 333 as 66mHz FSB.
T
1.the mainboard supports only 60 and 66 mHz FSB speeds.
2.the cpu(new) is multiplier locked and requires a 100 mHz FSB to run at 350 mHz.(3.5X100)
3.Compaq provides extensive coverage of legacy equipment like this and specific descriptions of the switches are available. Intel can verify the PII 350 is 100mHz FSB.
Very different cpu...from PII 333 as 66mHz FSB.
T
http://h200002.www2.hp.com/bc/docs/support/UCR/SupportManual/TPM_270866m2-ch7/TPM_270866m2-ch7.pdf
your system is set up to use the 66mhz bus clock while the 350mhx proc is supposed to be at 100mhz and if memory isnt failing me it should be at a different voltage than the 66 clock models
it is most likely going to be running on too much voltage(or too little) and be unstable leastwise.
waffle
your system is set up to use the 66mhz bus clock while the 350mhx proc is supposed to be at 100mhz and if memory isnt failing me it should be at a different voltage than the 66 clock models
it is most likely going to be running on too much voltage(or too little) and be unstable leastwise.
waffle
ASKER
waffle - thanks for the link (although I had that already).
tmj883 - getting really close to it. I now understand much better what is going on. Was that a typo, or does there exist a P2/366 with a 66MHz FSB ? or a P2/333 with a 66MHz FSB ? I couldn't find anything about the switches on the Compaq(HP) site, other than a chart for a few of the settings. Do you have a link that explains which switches control FSB and which are multiplier ?
tmj883 - getting really close to it. I now understand much better what is going on. Was that a typo, or does there exist a P2/366 with a 66MHz FSB ? or a P2/333 with a 66MHz FSB ? I couldn't find anything about the switches on the Compaq(HP) site, other than a chart for a few of the settings. Do you have a link that explains which switches control FSB and which are multiplier ?
Joshua7 ... there's no combination from the jumpers so you can run at 366, I'm sorry.It would need a 5.5 multiplier with a 66MHz FSB. The motherboard can't multiply with 5.5
As I already said, the speed increase would be insignificant, so I wouldn't bother is I were you.
KronosTM
As I already said, the speed increase would be insignificant, so I wouldn't bother is I were you.
KronosTM
366 with FSB not a typo...does exist but try to find one...and Compaq does list the cpu as having support. T
I don't understand your post tmj883
the only 366 out there is a celeron and wont fit anyway unless you get a converter card from fcpga to slot1, intel never made a pentium 2 366
though a 333 does exist and can be found for about $10
http://castle.pricewatch.com/search/searchmc.idq?cr=pentium+333&qc="PENTIUM"*+AND+"333"*+AND+ %40ctd+3&i =3&ct=Comp uter&c=Mic roprocesso rs+PC&mi=N &m=N
also after a bit of looking, i found that they both will run on 2.0v so you should have no prob there
waffle
though a 333 does exist and can be found for about $10
http://castle.pricewatch.com/search/searchmc.idq?cr=pentium+333&qc="PENTIUM"*+AND+"333"*+AND+
also after a bit of looking, i found that they both will run on 2.0v so you should have no prob there
waffle
ASKER
waffle - absolutely no instability whatsover using the 350 MHz chip - incorrect info
kronos - also incorrect. see what I post below for tmj
tmj - jackpot - I did some more testing and here's what I found:
at cpuid.com there's a utility called CPU-Z that gave me the following output
300MHz processor (came in unit)
SW CPQ CORE MULT SPEC FSB BUS
FFFF - 120 5.5 366 21.8
FFFN - 133 5.5 366 24.3
FFNF 210 210 3.5 233 60
FFNN 233 233 3.5 233 66.6
FNFF - 120 5 333 24
FNFN 333 133 5 333 26.6
FNNF 180 180 3 200 60
FNNN 200 199.8 3 200 66.6
NFFF 270 270 4.5 300 60
NFFN 300 299.7 4.5 300 66.6
NFNF 150 150 5.5 366 27.3
NFNN 166 166.5 5.5 366 30.3
NNFF 240 240 4 266 60
NNFN 266 266.4 4 266 66.6
NNNF - 120.6 5 333 24
NNNN - 133 5 333 26.6
333 MHz
300 5 333 60
333 5 333 66.6
350 MHz
210 3.5 233 60
233 3.5 233 66.6
What I gather from all of this is that the 333 and 350 processors are what they call 'locked' and are meant to use a 100 MHz FSB, and therefore don't respond to the switch settings, whereas the 300 MHz is 'unlocked' and can be made to do any number of interesting things. It seems that I have stumbled into that dark corner labeled 'overclocking', haven't I ? The only remaining question is which of the four settings which call the processor a 366 is least damaging, the low core speed of 120 with Mult of 5.5 and FSB of 21.8, or the high core speed of 166.5 with Mult of 5.5 and FSB of 30.3 ?
kronos - also incorrect. see what I post below for tmj
tmj - jackpot - I did some more testing and here's what I found:
at cpuid.com there's a utility called CPU-Z that gave me the following output
300MHz processor (came in unit)
SW CPQ CORE MULT SPEC FSB BUS
FFFF - 120 5.5 366 21.8
FFFN - 133 5.5 366 24.3
FFNF 210 210 3.5 233 60
FFNN 233 233 3.5 233 66.6
FNFF - 120 5 333 24
FNFN 333 133 5 333 26.6
FNNF 180 180 3 200 60
FNNN 200 199.8 3 200 66.6
NFFF 270 270 4.5 300 60
NFFN 300 299.7 4.5 300 66.6
NFNF 150 150 5.5 366 27.3
NFNN 166 166.5 5.5 366 30.3
NNFF 240 240 4 266 60
NNFN 266 266.4 4 266 66.6
NNNF - 120.6 5 333 24
NNNN - 133 5 333 26.6
333 MHz
300 5 333 60
333 5 333 66.6
350 MHz
210 3.5 233 60
233 3.5 233 66.6
What I gather from all of this is that the 333 and 350 processors are what they call 'locked' and are meant to use a 100 MHz FSB, and therefore don't respond to the switch settings, whereas the 300 MHz is 'unlocked' and can be made to do any number of interesting things. It seems that I have stumbled into that dark corner labeled 'overclocking', haven't I ? The only remaining question is which of the four settings which call the processor a 366 is least damaging, the low core speed of 120 with Mult of 5.5 and FSB of 21.8, or the high core speed of 166.5 with Mult of 5.5 and FSB of 30.3 ?
if you will look to my later post, i remarked that after further research i was mistaken.
as far as overclocking goes you want to go for the fastest fsb you can.
a system running at 100mhz is much faster than a sys at 66mhz
so while you are messing with the multipliers to change the "speed" of the proc, you are actually slowing the system down.
waffle
as far as overclocking goes you want to go for the fastest fsb you can.
a system running at 100mhz is much faster than a sys at 66mhz
so while you are messing with the multipliers to change the "speed" of the proc, you are actually slowing the system down.
waffle
Processor Speed Selection
The Pentium II-based system board includes a six-position DIP switch (SW1), of which four
positions are used to select the bus-to-core frequency of the Pentium II. Table 3-2 shows the
possible switch configurations for this system and the resultant core (or processing) frequency.
Table 3–2. Pentium II Microprocessor Bus/Core Speed Switch Settings
Table 3-2.
Pentium II Microprocessor
Bus/Core Speed Switch Settings
DIP SW1 Settings
2 3 4 5 [1]
Host Bus
Speed (MHz) Core Speed (MHz)
0 1 0 0 66 166
1 0 0 1 60 180
1 0 0 0 66 200
1 1 0 1 60 210
1 1 0 0 66 233
0 0 1 1 60 240
0 0 1 0 66 266
0 1 1 1 60 270
0 1 1 0 66 300
1 0 1 1 60 300
1 0 1 0 66 333
1 1 1 1 60 330
1 1 1 0 66 366
NOTES:
Shipping configurations are unshaded.
[1] 0 = Switch Closed (On), 1 = Switch Open (Off)
The status of SW1(2-4, 5) is readable through general-purpose I/O (GPIO) port 78h bits <3..0>,
allowing BIOS and/or diagnostic software to check an installed microprocessor with the switch
configuration. Table 3-3 shows the switch position-to-GPIO-to-I/O port 78h input wiring.
This is the cpu support table from Compaq Support and it is in agreement with previous assertions and CPUZ.
note: multiplier X FSB = mHz
Furthermore, it is not the FSB that is locked but the multiplier. With the multiplier on the 350 cpu locked at 3.5, on this mainboard with only a FSB of 60 or 66 mHz available,
3.5 X 60 = 210 or 3.5 X 66 = 233...
nuff said...T
The Pentium II-based system board includes a six-position DIP switch (SW1), of which four
positions are used to select the bus-to-core frequency of the Pentium II. Table 3-2 shows the
possible switch configurations for this system and the resultant core (or processing) frequency.
Table 3–2. Pentium II Microprocessor Bus/Core Speed Switch Settings
Table 3-2.
Pentium II Microprocessor
Bus/Core Speed Switch Settings
DIP SW1 Settings
2 3 4 5 [1]
Host Bus
Speed (MHz) Core Speed (MHz)
0 1 0 0 66 166
1 0 0 1 60 180
1 0 0 0 66 200
1 1 0 1 60 210
1 1 0 0 66 233
0 0 1 1 60 240
0 0 1 0 66 266
0 1 1 1 60 270
0 1 1 0 66 300
1 0 1 1 60 300
1 0 1 0 66 333
1 1 1 1 60 330
1 1 1 0 66 366
NOTES:
Shipping configurations are unshaded.
[1] 0 = Switch Closed (On), 1 = Switch Open (Off)
The status of SW1(2-4, 5) is readable through general-purpose I/O (GPIO) port 78h bits <3..0>,
allowing BIOS and/or diagnostic software to check an installed microprocessor with the switch
configuration. Table 3-3 shows the switch position-to-GPIO-to-I/O port 78h input wiring.
This is the cpu support table from Compaq Support and it is in agreement with previous assertions and CPUZ.
note: multiplier X FSB = mHz
Furthermore, it is not the FSB that is locked but the multiplier. With the multiplier on the 350 cpu locked at 3.5, on this mainboard with only a FSB of 60 or 66 mHz available,
3.5 X 60 = 210 or 3.5 X 66 = 233...
nuff said...T
Kronostm...366 cpu is supported and 5.5 multiplier is available according to Compaq...see above...T
PII 333(66) avilable at:
www.pcprogress.com
US$12
cheap, but I cannot find a 366(66) available anywhere...T
www.pcprogress.com
US$12
cheap, but I cannot find a 366(66) available anywhere...T
all i was saying above is that if you head to www.intel.com they say that they never made a 366
i can see what you are saying though, if the board in fact supports a 366 it should have a 5.5 multiplier!
waffle
i can see what you are saying though, if the board in fact supports a 366 it should have a 5.5 multiplier!
waffle
For a definitive article on the Pentium II please read:
www.dwpg.com/content.php?contid=2&artip=89&orgartid=89
This article will explain EXACTLY what to do with your processor...hope the link works...T
www.dwpg.com/content.php?contid=2&artip=89&orgartid=89
This article will explain EXACTLY what to do with your processor...hope the link works...T
Sorry, the link comes up short of the article...use the "can't find" search by typing "unlock" for a good link to the article. T
there is no 366MHz CPU for Compaq Deskpro 4000 6300MMX. Let's see if the finalisation of this Q will contradict me :-)
... listening .....
Switches 2,3,4 set the FSB(for a multiplier of 5.5 the setting is 1,1,1: the final switch 5 sets the FSB, 1 for 66mHz) Too bad the 366 cpu only exists on paper. T
2. The cpu is running at a slower speed because the mainboard does not support 350 mHz cpu.
It supports 233, 266, 300, 333, set by the DIP switch. The cpu is defaulted to the slowest speed as it is unrecognized.
3. The BIOS is telling the truth...it does not recognize a 350 mHz cpu and never will.
4. As explained above...the 300 mHz cpu(original) will operate faster than the unrecognized 350 running at the default 233 mHz.
T