Silent Antec CHEAPly

Show off your quiet rig.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

July 1 update UPDATE

Post by colm » Sun Jun 21, 2009 8:56 pm

Image

big day for the 6 year old antec 2600 amb. I took 10 minutes out to paint it, in the seemingly 1 hour of sun I have seen in the past month. (my locale has a record this jun 2009)

1 year paint code for a 1987 subaru. the paint is 22 years, the pc hours are at 21020..next update fall 09. note the pink and bluntly brown..it is a dual range paint for a dual range subaru. It looks lousy right now, will squirt again once I get to my car chores with real urethane and fine sanding.

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

Post by colm » Wed Sep 16, 2009 11:17 pm

Image

it is before fall, but 37 degrees, headed for the fulfilled freeze warning this AM in my location is before fall weather too. :cry:

therefore early update. I have the main cpu fan making a bit of noise sometimes, and may change it out. Other than that, all is well. I did resort to mozilla firefox (I haven't veered off iexplore in years) internet explorer is doing something uncurable, also not seen in years. I am on an 8mbit connect, maybe something is overdoing it.

I was a volunteer for the dimaond thermal compound, and am impressed. my outside weather is my inside weather, and it is holding strong even with the invisible chill creaking through external walls...just a few weeks ago, it was the smells and clean rains of a leftover hurricane. 75F at midnight...

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

after fall

Post by colm » Thu Sep 24, 2009 9:09 pm

Image

for logging purposes, I forgot the volts temp photo. The locale this time of year is headed for brutal on anything machinery. In fact I made a part for my car that does not exist, just for weather extremes recently. This keeps my eyes open to the temp and volts above. A new vid card driver and the system temps have been 5 degrees higher. I tested the vid with steams stress test in video games, the new avivo encoder has the hd option, and it does very well very fast... a 350mb mpeg2 to hd in about 1 minute.
I have no other benching that I do aside from encoding. The system temps higher may bring another cooling change. I am not sure what yet...

next update december 25th. prediction of making it to 25000 hours.
I have an 18 hour leeway, if I do not shutoff the computer for the next:

From and including: Friday, September 25, 2009
To and including: Friday, December 25, 2009

It is 92 days from the start date to the end date, end date included

Or 3 months, 1 day including the end date
Alternative time units
92 days can be converted to one of these units:

* 7,948,800 seconds
* 132,480 minutes
* 2208 hours
* 13 weeks (rounded down)


a little challenge for myself. my current uptime is only 2 hours and 40 minutes

92 days to go. I am one who had windows millenium up for over 48 days...win2k up for 120, the vista is win2k type, I think it can do it.


Until next post... :)

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

SERIUOS UPDATE

Post by colm » Thu Oct 08, 2009 9:56 am

I hardly interrupt the equinox posts, unless there is something important to log.
I burned my ati 2600 pro. It did last over 10000 hours, it died around 23065 hours, the longest in my ati history of 10 years and a dozen cards. The ati 2600 agp had some difficulties, but custom cooling inspired by SPCR kept me enthused, and well, the most difficult card ever lasted 18 months. I cannot complain.

Image

here is the new card. It is a visiontek ati 3650, fan power managed (very nice, it slowed down to <1000 rpm as of this moment) . It still keeps windows index rating at 4.4, the volts are down to .90 and it is a 10nm drop in size from the 2600s 65nm. I almost crapped when I saw it had no heatsink on the pci-e convertor. They get very very hot! So I dug out my old athlon 900 heatsink (socket 462?) and cut a nice fat square chunk out of it. and gave it some diamond ic compound, and the soft layer of foam looking stuff got some 3000 psi epoxy. Instant transfer. The top temps after testing on the card are an mazing 114 degrees, the idle is the same as the 2600 temperatures, around 39C.

the new fan I purchased, a dell remnant with a thermistor built in may have been the cause of the ati cards demise. ta300 nidec.The fan had no function what so ever, and I took it apart in rage to find it had the highest grade sealed ball bearings I have ever seen. they seem to be encased in a plastic, inside the steel casings...filled with fluid. wow. where to get those without the bad fan to go with it? anyway, I built another fan with the new bearings, a new one that had bad bearings from the get go, and so far, that too is the quietest low overhead fan I have run to date. It was a matter of the right sealed bearings..


If this makes it...I will be posting again Dec 25th, still a goal of 25000 hours. I cannot shutoff the machine again for any more than a couple of hours to make the goal. :oops:

some testing notes. with hwmonitor (catalyst is stuck, can't even untsick it) I used a benchmarker for video by freestone group (basic stresser) and learned the power management is as follows:
110mhz == .90 volts
725mhz == 1.25 volts
and the 55nm is proving that 725 mhz is the same max temp as 601 mhz (2600 pro) at 65nm, which is 53C after 24 minutes, and the same for hl2 for however long. it doesn't go beyond it.
and the idle is 39c.
There is one oddity in the 110 mhz mode (power saver) it kicks in browsing around the net, and even stays there watching net videos... old sir prescott could care less about a video card. like a blast from the past of ati attempting this, the sidebar has wiggly lines through it, hardly noticable, but I do. The rage lt pro also had power management...the year? 1997. :roll:

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

update again

Post by colm » Sat Oct 17, 2009 12:28 am

I posted this in a new thread in the video card section, I have a habit of letting this one go for a few months at a time. Unless something significant. Well, I ran into a problem with new vid card. I was getting wiggly lines on the edges of screen, and guessed it to be heatsink without compression. the accelero is just right, so I concluded it to be the pci convertor chip. this is what I came up with, it is snug, and it is dual purpose. It is helping the accelero and the gpu stay snug as well, round about way.

Image

the metal is aluminum, bottom side painted, it is an exact fit. All four studs in the gpu cooler are my own mod, longer and stronger machine thread. the mobo studs are used for easier install of the pcie cooler. I noticed the accelero took on a perpendicular a bit more correct after this maneuvre.
this fixed it. top temps are 45C in an 80+F room. off to 25k hours..with a bit of luck. As of this post I am at 23337 hours.
I did need to reinstall vista. I have it setup in a way where it is a quick install. It seems the vid cards are complicated enough to justify it. uninstalling drivers did nothing for cleaning up the registry. I got over a year with vista premium no problems. ATI and thei card uninstall is not an easy task and even bothers everything. all worth it somehow. :)

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

Post by colm » Sun Oct 18, 2009 10:17 pm

jack34 wrote:Thanks for the useful post
certainly. It is a vow to return what I learned the hard way. I found this place nearly in the midst of a small stroke quite some time ago...in several thousand dollars of absolute junk. :roll:
anyway, always enthused..kinda...

another tidbit I added to the vid card..

Image

again I interrupt the equinox schedule. I noticed the high end pci-e all are extremely covered. So, with this ugly piece, I wedged it into the pci a heatsink (previous post) fin, and left an edge more inboard open than aft, and it hangs over the edge of the card. It is actually an act of precision, like the ugly duct. My 2600 had all aluminum, that was not good. This plastic is simply to ride a flow I know is happening. the vid card temp is not cooler, but the system temp is now down a few more degrees...it is working.

Image

here is a view of the snug fit of the plastic (thanks maxwell house)
all memory is covered, and everything else of course that is underneath it. the temps are exactly the same, except for quick cool downs..they no longer can happen. My locale needs this as much as cooling: a soapstone effect in the winter rather than a cold snapper.

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

old sir prescott passed away

Post by colm » Sun Nov 08, 2009 12:12 am

Well. big news.

after 23900 hours rounded down, the motherboard lost a vrm while foldiing.
Not one to get discouraged, it is the best run I have ever had. I removed all components, and await my next build. Right now the "silent antec cheaply" is hosting a 633 celeron (circa 2001).
I decided to paint the case. The first coat is from an 87 subaru color code (previous post), its a duality paint, pinkish metallic and plain ole brown. I decided to add some copper flake, which is not real, it is polyester and aluminum made to look like copper flake. I concluded silence (polyester) and radio deterrent (aluminum). This battle (for me it is a battle- I gained a psychosis about a decade ago, related to computing), is not much of a battle anymore. I enjoy computing after this long run..and it is not done.

Image

here is a glimpse of the case. two part automotive urethane is two coats. one coat holds the flake. note it is not smooth, intentionally so...





overall stats:
73% uptime, over 3.7 years. impressive. I make mistakes that went for the ride.

the case is holding its 4th mobo as we speak, the orginal power (sl300 smart power)is back in it for the celeron, will be going back to the earthwatts 380 for the 875pbz.

stay tuned for next build...

Image

here is a photo, out of the case it sat contently in for...well since just before the first post here. Not to be sentimental over a machine (but I am a former crew chief)...I couldn't help but pause for a moment and stare. Many hours for one person and a machine. many personal events, many serious chores, purchases, and some code, and over 300 videos. A first for me in this longevity, and with enthusiasm, certainly not my last.


EDIT: I found a large supply of this motherboard, and will be replacing for nearly the same cost as 4 years ago. To me this is the first batch of builds in history (there aren't many) , where you can fix it, and still be content with performance. so, I will be a few weeks with this antique celeron, and then back to normal if all goes well. I literally tried to fix the vrm that hissed (no poison, just simply done), the board is still as shiny as I put it in, also a first for me. That ius ashame not fixed yet apparently, about vrms..once it goes, there is a great odds to keeping it going with repair. A complete lack of recycling ability...to this day... :roll:

YET ANOTHER EDIT:
I found a rather serious error inside the earthwatts 380 psu. A large inductor, supposed to be wrapped with a pastic cover, had a free floating plastic cover inside the psu and no wrap. The inductor design is very soft copper, and was allowed to move wiggle and even vibrate without this covered.I risked warranty (as usual) and removed cover and zip tied it to the inductor in a way to keep it still and protected, better than their oe attempt with what looked like hot glue.. the power on (power ok) signal is 100% brighter...
This is a means for a babbling argument with antec support, but I won't bother...
and with enthusiasm, the case is on its 7th coat of paint, 4 are urethane. It looks deep.
I have concluded: it is covered.
I found some scraped down spots on the inside cover, and smothered that with one coat. The urethane is top grade and expensive. I play with half cup mixes and less at a given coat.
Awaiting for rebuild in coming weeks, with same board. With spots I found also needing covering on the board. Empty raid spot, and underneath empty vrms (yep, underneath?!). that should conquer a few more anomolies. This time of year is a pc killer where I live. Right now is 58 degrees for example, and tomorrow morning in the 20s. I have had cases pop shape to strange noises and movements, even in a stable room. The temps are dramatic.
This very scenario survived several years, with scratches and a crazy psu this past year, it should be better next go.

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

Industrialization has begun

Post by colm » Thu Nov 19, 2009 7:59 pm

Industrial strengthening has begun

I bought a camera that passed military shock standards, and for years have wondered why motherobards did not get treated the same...why?
It is only a layer of epoxy/urethane, sealing in resistor ends, empty chip spots, and insuring grounds are grounds and no bleeding allowed.
motherboards and chipsets evolved too fast to even bother with making tougher. It has occured to me, even big business has hung onto a run of socket 478 cpus, and I have decided to go some extra steps...

The ureathane is specifically "xtreme", and I could not believe my own eyes at what torture that took on in my awd subaru sandblasted in the high 70s mph in the dead of paint cracking winter. It is the reason I chose it...it even took on a battery type chmecial that gave me a nosebleed.

Anyhoo, the chores, I do not recommend this for just any diy:
  • poured urethane on the bottom of circuit board of PSU, and verified points weren't overstepping boundaries.
    quickly light painted the inside of case, urethane. remionds me of n older case, it is a "once upon a time they had urethane" story after all..
    the outside of the case had respectable oe paint, but refer to previous posts at what I have done. I stopped at 7 coats, 4 are urethane. the non colored (gray) also got a quick coat..the whole case in other words.
    stainless steel backplate is too lively (refer to the past of thick back plates, non-conductive) I urethaned that as well.
    Verify front cover connectors, and wiring that came with case. My own 2600 AMB has a small circuit board for usb, making a smooth transfer. I simply urethaned it.
    Air filtering, if I have not mentioned, is done with an aluminum screen, and a cover from a cb radio (SECC only!) to maintain consistency for hdds.
    Reverse cooling, urethaned custom duct, no volts, conductivity allowed.
I think this covers everything, pun intended. Staying well within specs, I could hear the earthwatts fan ramping up for no reason, while pc idled..it was responding to the very air getting sucked into the pc. No more. I have lost for the last time to unfinished , lack of quality assurance...
I will also, as mentioned, be urethaning the motherboard...
ponder a decade, and 5 grand in pcs, just for the internet, before seeking my qualifications...
maybe 50k hours is not that unbelievable. If a wd hard drive can do it, so can my build.... :lol:


my track record with failures, has proven my "stuff" cannot handle temperature changes , nor what happens to electricity in those temperature changes...and the odors. oh my...
So, I knew urethane worked on everyhting, with a 340C boiling point, I searched the web for a pro quote that identifies what I found by using polyurethane.
Many electronic components and assemblies require a potting compound for protection against moisture, thermal shock, vibration and very low temperatures. Most of the potting & encapsulating compounds available to designers are unable to give adequate protection over a wide temperature range. Epoxies are generally hard and tough materials. They provide excellent mechanical properties, good chemical resistance and good adhesion. However, they do not protect delicate components during thermal cycling and may provide stress during the exothermic cure. Silicones are excellent for high operating temperatures (185º C) and provide a low durometer. Unfortunately, they are expensive and do not provide good adhesion. Conventional polyurethanes can be formulated to be flexible or rigid and have a low peak exothermic temperature during cure. But, these conventional polyether and polyester urethanes can be penetrated by water
the last sentence is an oxymoron. absurd...

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

prep for new board continues

Post by colm » Sat Nov 21, 2009 6:38 pm

I took everything out of the case again, and slowly looked around....
I spotted the screw hole openings for other cpu retainers.

The prescott has none...my board is atypical anyway for a 478.

looking over the motherboard I noticed even the memory slots bent over the hours of runtime.

so I added two studs, the "south" ones of retainer near northbridge. what a freaky game of exacts that is! I even had to mount the stud exactly in a way as to not touch circuitry on the specific board I am using....less than 2mm..
Motherboard tray non-removable (I love it), I still found a way to slide a nut underneath and lock down what I believe to be the most stubborn stud in the future. The northbridge heatsink is stiff, after doing this.

So now retainer is bolted right to the board, fine machine thread, in two spots anyway...
It exactly counteracts the bends I found. This does work, as my first pc, a slot 1 pentium 2 stayed running even after the southbridge touched the case...the distance the studs are committed to assure runtime, and even aids in errors.

Next post, up and running, hopefully. A 5 year old board getting aced in a case it proved itself in, even slightly flawed, is worth it ...


oh, I did urethane the bottom of psu circuit board, very thick, and ran it after 24 hours on another motherboard. Nothing to notice... except the fan seems like its quite "solid", if you know what I mean... ?


:wink:

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

urethed antec 2600 amb

Post by colm » Sun Nov 22, 2009 7:53 pm

Image


I caught a glimpse of the fuse box near me. grey. very hard enamel....
and let some cleaners go to work on the 2600 case for awhile, then applied some krylon grey, identical to fuse box, then I continued on the next day with 2 part automotive urethane. It is by far the nastiest paint job I have had yet...it REALLY kicked something violent out. In fact I have a sunburn putting my face near it. upon the story of 3 mile island leaking...well,, I'll try not to let my facts fly..woops, I mean imagination.

8 hours later, I could hardly stand to be near it. It is finally calming down. I did add some 200 degree heat after 4 hours, (a hardness trick), may as well call it baked enamel.

I added two studs specifically for the ms6728 motherboard. I hope the places I found them for sale do not have a database error. I built the case counting on a supply of the motherboard. Will find out tomorrow, if to share some anticipation.

if not, I did find an 865 perl, intel board similar, except my two new studs will have to be removed.

KansaKilla
Friend of SPCR
Posts: 381
Joined: Fri Aug 01, 2003 12:13 pm
Location: Rochester, MN

Post by KansaKilla » Mon Nov 23, 2009 12:32 am

There's nothing to say here but...

WOW.

Hats off to you, colm. I mean that.

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

Post by colm » Mon Nov 23, 2009 10:10 am

KansaKilla wrote:There's nothing to say here but...

WOW.

Hats off to you, colm. I mean that.
I hope anybody is inspired to do the same. I am not even making a showroom worthy finish. I lose and lose with electronics...all except for the explosion proof military camera...
Some years ago, I began asking questions. Even top brand stuff is half-a**ed more than surprised friends etc could guess, in blindly with a chunk of change all at once for a pc that exchanges emails (newb users)...and losing..sooner than later..very frustrating..and their is always something annoying about noise coming from incorrect systems...

It really sincerely got on my nerves...literally physically sometimes. My difference was starting with a trs80 that exploded, in 1985, after being part of an experiment in the 5th grade with a computer called an "apple" (1983 or 2?). I learned right away we are all experiments apparently for something too new to nail down exactly correct. As most know ..it seems near anything goes to this day, and it is a huge clock now...

silence starts with correct cases...I am certain the boards and chips could go many years, as some of my other stuff as proven. A 1983 rca tv I have still runs, it ran for so long the glass is yellow in the inside (non fixable)...another clue to attack the case for the pc.

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

Motherboard purchase decision

Post by colm » Tue Nov 24, 2009 11:17 pm

Bad news for replacing the platinum. I found just ONE worldwidesearch..and it is used.
I did say silent antec cheaply and did my "classic" homework

I found 1311 d875pbz motherboards in stock, brand new, and there is a branding of "gateway". No biggie, I verified that they had the latest bios for download, and raid. If I am not mistaken, the 875 was running raid on ata and serial, and not advertised to do so. Will wait and see..(some 865 had a surprise for some owners as well)

as far as upgrading, I simply stepped up to a board with no onboard sound.
The thought of intel raid, and intel lan with intel chips has always been a favorite (don't tell anyone) :)

And for some five year old facts to go with the board..

it will take the 3.4e cpu, of course, but it will be under 4 phase power, the platinum 865 only had 3, and more than ironically had an empty space for this and never used it. So, popping a mosfet on the motherboard holding the biggest 478 cpu may go away as a problem. And, lastly, it has a thing called PAT, as if DDR ram wasn't the lowest latency highest speed on earth already, pat tightens this even more by reducing latency....

the 865 and a double dual channel test with HT enabled was 6.4GBIT bandwidth ... I wonder what the 875 does by dropping latency...

The next post is putting together, and the intel boards have actual mounting stud holes at the retainers. Good news.(the 865 had a circuit less than a mm form retainer hole!) One more mod on the box, a quick urethaning and away it goes, hopefully.

if anyone is in need of a 478 board,here is link

edit: the link may be for america, or north america, I do not know if it is international.

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

back to 865 perl

Post by colm » Fri Nov 27, 2009 12:54 pm

Upon further study,
The 865pe chipset choice almost 5 years ago, is going to be my choice again. I liked it enough to perhaps go for a bulk buy (seriously).

9 more standoffs and 23 more screws , and the case will finally be ready for the infamous word "refurbished". I could believe it for a pc case before I wll never trust it for a motherboard for sale.

the board will be new, with 14 standoffs!

anyhoo, the details seemingly never end, and then taken for granted later on when it seems nthing is going to go wrong. It is a rude reminder everytime.

so, in conclusion of decision (finally...previous posts have several final answers), it is transeferring to perl, essentially trhe same only better, and thorough documentation by intel reveals what I have built is as good as it is going to get. Also the xp90 will be hanging over the vrms, (turned sideways) with a new duct for reverse cooling, also advised by intels docs to have alot of air.

and a new duct. I hope it is considered clever...and cheap.

this will be yet a few weeks, but to ponder locally...I paid 130 bucks for an 845g some years ago..( and they were done already) and it took no ram from brand new. that is my local for sales..can you imagine? The net did reveal these are just plain bad builds, one of many, like my platinum missing vrms ....
I would swear this place needs a reseller...in reality. I said it ten years ago and "may get off my ass" and do something about it....

there are old celerons with dx10.1 vid cards in a pci slot, on 66mhz busses and 512 ram just as content as the day they started....the stampede out of the way is a great place for a reseller to start, by finding real longevity of real computers with real people.

Anyway, thanks to the net and oem overhead of large slothed companies. I get to continue with the prescotts.

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Fri Nov 27, 2009 3:32 pm

Amusing and interesting thread, Colm. Thanks for keeping us uptodate on your journey. ;)

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

Post by colm » Wed Dec 02, 2009 3:12 pm

MikeC wrote:Amusing and interesting thread, Colm. Thanks for keeping us uptodate on your journey. ;)
hey thanks.

when I found a fried nvidia at one relatives house, an "antique" celeron 633 still running at another, a dead p4 in more than one place... no enthusiasm to this day by any of the owners...

all associated with case, power and cooling....

the wavelength of actual evolution hit me really hard...yet again, it is a reminder to work with what you've got. (I am here in a 9 year old pc in proper power, case, and hard drives...)

I did find a brand new, not refurbished 875pbz with sound, yet again I have changed my mind, and am going for this.
I recently remade a duct, much more curve, emphasizing static pressure (yes in reverse) for the mosfet cooling mentioned in the intel documents.

it is a simple experiment to find when it is good.
also, I am attempting raid, and have a small heatsink for the 82801 er chip...

also modifed the large sink on the ati3650 (the top one) now it is smaller, ironically a cooler from an old apple ati rage...bolted down very presicely this time, no wiggling around.

stay tuned. oh..it is all on 380 watts.

I cannot take photos with this pc and trust integrity, I wait for the newer density to be up and running. that is actually the only reason ...the micron changes photos appearance as well as upload to pc. why chance it...

Monkeh16
Posts: 507
Joined: Sun May 04, 2008 2:57 pm
Location: England

Post by Monkeh16 » Wed Dec 02, 2009 4:03 pm

colm wrote:I cannot take photos with this pc and trust integrity, I wait for the newer density to be up and running. that is actually the only reason ...the micron changes photos appearance as well as upload to pc. why chance it...
I'm just trying to figure out what you're saying, here.

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

Post by colm » Fri Dec 04, 2009 12:37 pm

Monkeh16 wrote:
colm wrote:I cannot take photos with this pc and trust integrity, I wait for the newer density to be up and running. that is actually the only reason ...the micron changes photos appearance as well as upload to pc. why chance it...
I'm just trying to figure out what you're saying, here.
its in plain english, maybe you can learn it some time. high micron cpus cannot show a photo correctly. It did not go normal for me and my nikon until the prescotts. I hope this explained a little more.

updates:
as no vid card is ever normal, I added a few more things. The pcb in card is not a thick one, pushing 725 mhz. Alot of cards have a backplate, but it is multipurpose, all aluminum, and I don't like it. This horrifying visiontek got shipped to me "brand new unopened" with slime like vaseline, missing a heatsink. So here is a seperation of pci-e convertor cooler, and SECC around it pushing on the back of GPU chip. I am not up and running, this is just some of the chores moving forward

Image

I have successfully swapped out the earthwatts 380w oem fan for a zalman equivalent (.25, and reporting) I will grab a photo of that next. I have it set to plug into the vreg spot of 875pbz, as it does report it to a software monitor, and of course stay powered and throttled by its psu plugin, not the mobo. that is just a reporter (two wire hack, like a 300w "smart power" used to do) As it turned out the oe earthwatt fan barely survived a burn in, or whatever they do with them. The noise was bizarre..the zalman fixed that as it came to life at very slow rpm, capable to near 3000.


I have read more specs on the pcb for the 875 chips, and they are six layers. I am still mounting mobo to 14 studs in all (the older are four layers)

and a bit of my own review: I am done with buy.com

Monkeh16
Posts: 507
Joined: Sun May 04, 2008 2:57 pm
Location: England

Post by Monkeh16 » Fri Dec 04, 2009 4:11 pm

colm wrote:
Monkeh16 wrote:
colm wrote:I cannot take photos with this pc and trust integrity, I wait for the newer density to be up and running. that is actually the only reason ...the micron changes photos appearance as well as upload to pc. why chance it...
I'm just trying to figure out what you're saying, here.
its in plain english, maybe you can learn it some time.
It's really not, and considering I learnt English where it originated, I'd say I know it pretty well.
high micron cpus cannot show a photo correctly.
They... what?

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

Post by colm » Fri Dec 04, 2009 8:19 pm

Image
click on thumbnail for full image.

Gathering some more tidbits of fact. I forgot to post the photo of the xp90 heatsink. This failed on many prescotts. This one started bending back when I had a 2.8e installed, almost four years ago.The arms on the outside simply bent leaving the cooler completely ineffective, and of course once they start to bend, the plastic retainers have a chance to break.

Not wanting to give up, back in 2006, I found a two part epoxy, the toughest version, at 3000psi in the auto store and added some small aluminum scabs ... something electroplate type physic happened, as this thing never budged again, right into 165 degrees fahrenheit....for DAYS at a time...Something tells me this could go ALOT farther than that...

Next photos are the Antec 2600amb errors I encountered and fixed easily. SECC and errors is better than something else claiming good... :)

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

Post by colm » Wed Dec 09, 2009 8:06 pm

as I have had the antec 2600amb case apart, I took some shots of the mods. very basic, nothing beautiful, extremely necessary.
Image

the fan hole can now do 80/92/120 mm fans, I did this quite some time ago.

Image

I finally whipped out some rage on the io shield. the 875pbz going in next does have a more solid version that I am used to, and I went some extra steps with secc, gray primer, two part urethane, and some rivets. Now it IS an io shield. the crazy duct underneath is taped for a dry fit, will get even more rivets, aluminum and paint if it fits the way it is supposed to.


Image

this is the front of the case, it has needed one rivet. I spotted it several years ago. the .8mm secc steel even without paint has proven to be clever in this unibody cheap case.

other mods include making the usb wires into a jack as a normal one, instead of a bunch of individual pins. It is a standard after all. I am not sure why antec left these all individual, quite a nuisance.

if all goes well, I should be up and running after tomorrow.

the changes to look forward to...

the fan and temp control is all lower level than user intervened sofware in the os while running. this means, no crazy ups and downs, it is in a function on its own, at bios/chip level. I just get to read only the temps and speeds.

the psu will have a reported fan to the system, as if that isn't a standard that should never be a luxury, but necessary.

I did read the 875 is workstation capable, that means intel did not "play" with this board, as me and my history with them has proven. ECC capable, and thier is a function specifically benefitting the 800mhz FSB prescotts only...

I hope I do not need luck. This takes some time, and to learn your new mobo is bad, well, it is a frustration to say the least. I did order new, and hope all goes well...

I have had 3 bad mobos from the get go, never started from day one out of 12...that is indeed 25% fail.




:)

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

875 up and running

Post by colm » Thu Dec 10, 2009 4:50 pm

Image


well, I am up and running again. the picture was taken quickly. A glimpse of a rather large northbridge heatsink ....

several things to notice...

the fan control is sucking as of right now, but at full speed quieter.
It must be the half a quart of urethane in the tiny case, and 7 layers on the outside.

there is ECC checking whether I like it or not. My memory is not ECC, yet it has detected an error at boot (I shut down midway booting, and it told me there was an error :)
and it refused a file from the net claiming corrupt... (nice. Haven't seen that truthed in years)
and lastly, the wndows updates take freakin forever to install...it is checking every last inkling of a morsel apparently....

the result is hellacious fast. It is like flicking through tv stations.

I give it a day, updated bios is last one for this board (the p34)
and it started no probs with the large prescott, before th bios. The board is a "209" black revision 2. About as new as they got.

a new duct, flexible, rubbery, I conformed it to where I wanted. the curve may also be the new silence.


and this board helped me dicipher my burnt one..the ms 6728 platinum was a lunatic from day one. That is all I have to say. Several errors, the most obvious the VRM array (simply absurd)

AND another and...

the vid card on slow mode (110mhz gpu and .8v) has no wiggles on the CRT. It is either the SECC backplate on the card, or the 875 is simply smarter.

and the 875 is synomous with a certain server chipset. (I found it in device manager.)


I hope this leads to another long run.

hd tune tells me the drive is now at 24248

this is the new zero.

the extra motherboard screw (there is 10) and the integrity of pushpins (decent) by oe allowed me to change my mind about adding studs under the cpu. It is all an oe mount as of now.

I do have one problem with intel active monitor. It doe snot work. the bios reports the psu howver in the cpu plugin, and the front and rear fans as well.

the cpu read 195 degrees for no apparent reason, I guess it may be a trait with the 478 run of systems.. i have not quite seen a correct one.

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

speedfan works 875pbz

Post by colm » Thu Dec 10, 2009 8:35 pm

Image


After 8 hours of no calming the fan, I resorted to a program I donated to, 5 years ago, called speedfan.
This scenario with the 875p really made my 8th hour of uptime.

rheostatically the fans have many positions in throttle, including cpu, whuch can be seperate. I have what you see in photo. And the cpu temp is way way off, it is cooler in this 875 than in the 865, and the reading makes no sense whatsoever. Speedfan allows me to work with it.

Things are going together well....

EDIT:
I did find an anomoly I haven't had in years:
The thermal compound had to be just right. sorry to say the diamond IC stuff had to go. I put in some thermalright stuff that came with my heatsink, and did not need speedfan. the fans all calmed down instantly from the second I started. I was hoping for this resolve, as any software linked to the bios that does the regulating goes a bit whacked out, no getting around it. It is ok just to read speeds and temps, but an ability to software it has never been right for me.

and a note on the psu...
the fan has not gone over 2k rpm, and it is very nonchalant at full load. the urethaning was extremely necessary. The io shiled is a bullet proof vest if to analogize, the sound is clear and precise. This bings a decent future guess on the vrm array and longevity...

I hope to give this thread a couple of week=s or even moths off to a new setup.
I do recommend the 875, late revisions only, if you are troubled by the hot prescotts, weak retainers and crazy boards that many many came with. the fact is, the prescotts are the predecessor to today...and it is in the worlds most efficient socket desgn, called pin grid array. Not may excuses but everything that surrounded them, to make it not what it is, and what it is, is damn good.

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

temps screenshot running

Post by colm » Sat Dec 12, 2009 1:20 pm

I finally found a fantastic monitor, I had forgotten about.

this is motherboard monitor version 5 something. I first tried this about 10 years ago on an impossible motherboard.

Image


I let it run overnight , the frigid chill permates our houses for the next couple of months. I have even found the front fan to slow down seperate from the cpu/case fan, and the psu fan goes no farther than 2300 rpm. Impressive intel integration...and the icing on the cake was the fans regulating in a dos type mode for a memory test, no software needed, when system temps (north bridge) can hit 120 F in such testing, it knew to regulate without me. Beware crazy motherboards, this 875 is not one of them...and I will never call it the past.

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

vid card bios

Post by colm » Tue Dec 15, 2009 9:03 pm

As I am one to leave well enough alone, a minimalist, the following photo shows the scariest thing I have ever done to a pc....



Image
This photo is from techpowerups rbe program, for ati video cards,it can look at bios settings, and write a change using the same signature as oem.

vid card slow mode (110/252 mhz) problems:
  • My absolutely serious problem was blurry screen with oscillation waves at 110/252mhz in slow mode. I was going crazy at slow web pages with heavy graphic...

    and how does the mouse cursor get affected by an ati card? .. the slow mode even stopped from typing too fast..and I type with two fingers. the flash fixed it.


I was nearly going insane until I learned I could full throttle it with no bios flash, or flash it and half throttle it by ( in photo)...and so I followed instructions and wrote a new bios, and it flashed succesfully. So now in what was supposed to be 2d mode at 300/400 REALLY is 300/400 mhz. Why anyone would want 110/252mhz is a bizarre answer I am certian. I could not even open heavy web pages with out thinking my 633 celeron with onboard vid was doing better. It seems I must have had a mistaken card yet again. Interesting note is that the chip is genuine ATI, that made my day.

The changes obviously are written and it does change speeds by screen appearance obviously to my own eyes. ATI catalyst shows the 2 modes. HIGH and LOW (photo) And I do not recommend this task to anyone, just return the card if it does the same as mine did. It can destroy a whole system. A "bios puke" can do some crazy freakin things. Like micro bulletts.

my own card is void of warranty already it seems as the heatsink took it over. Interesting note is the vid card fan does not begin to throttle until 50C and the card with the accelero never sees 50C. That means the fan is in slow mode at all times, which is good, I do not have one plugged in.

another means to keep quiet has been fixed... I could have stayed full throttle, but the screams would emerge eventually...I like this dynamical approach. :D

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

heat shrink tube on fan wires

Post by colm » Sat Dec 19, 2009 1:10 pm

straight from hong kong to maine (Distance is 12,619 kilometers or 7841 miles or 6814 nautical miles). I must have my heat shrink tubing. 8)
, it took awhile...(customs).
heat shrink tube, 2.5 mm for wires. I got 9 feet of it for under 3 dollars.

It needs 125c to shrink it.

So I added some to the psu reporting pair, right over both with one tube, the front fan near the connector, all 3 wires, and my back case fan/cpu (same fan) also all three wires, pairing the ground and positive same tube for most of it...and then shrunk it all.

I wasn't expecting to notice anything.. until I saw cpu volt up by .03, at full load.
The front fan gained just under 200rpms at full speed, the back fan gained 236 ...
The psu fan seems to report more dynamically, the speeds change more than I thought.

all this with heat shrink tube on the skinny fan wires.

..this deserved a post of its own. :D

anyhoo, the further plan to keep the rest of my teeth that did not explode from the last involunatry pc electrocution (thanks compaq, its worth a million).

here is what I did to the reputable safe earthwatts 380w

Image

the splice are covered in heatshrink tube, good for 600v, the fan a zalman. The oe fan must have been bad, the zalman in the same circuit stays very calm, and a very broad range of rpms. the .25a matches the oe ADDA fan.
to get it to report, I simply shared the fan ground and sent the reporting wire out together to the plugin for the motherboard.

the hazard with this supply was the freefloating lightning launcher at the top, an uncovered inductor. I strapped in the plastic u-shaped piece, and again, the system calmed way down at full load. I am beyond assuming it was going for a ride to the lively heatsink on one side and the ciurcuit board on the other. Thanks antec, this error is only worth 100 dollars. No worries, the motherboard was a "platinum edition" rare, and is extinct, I have moved on.
I have also urethaned the heck out of the back of the psu circuit board, another bizarre lively world of make believe out to rest...

have fun kids. before the pc-pocalypse steps in. :lol:

edit:
forgot the close up photo of the lightning launcher at peace. A millimeter is indeed a millimeter.

Image

this was instant cure. as the photo reveals the bad glue seam.
Last edited by colm on Sat Dec 19, 2009 4:36 pm, edited 4 times in total.

Monkeh16
Posts: 507
Joined: Sun May 04, 2008 2:57 pm
Location: England

Re: heat shrink tube on fan wires

Post by Monkeh16 » Sat Dec 19, 2009 2:19 pm

colm wrote:So I added some to the psu reporting pair, right over both with one tube, the front fan near the connector, all 3 wires, and my back case fan/cpu (same fan) also all three wires, pairing the ground and positive same tube for most of it...and then shrunk it all.

I wasn't expecting to notice anything.. until I saw cpu volt up by .03, at full load.
The front fan gained just under 200rpms at full speed, the back fan gained 236 ...
The psu fan seems to report more dynamically, the speeds change more than I thought.

all this with heat shrink tube on the skinny fan wires.

..this deserved a post of its own. :D
Quite how you think that's anything to do with the heatshrink is beyond me.

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

Re: heat shrink tube on fan wires

Post by colm » Sat Dec 19, 2009 3:23 pm

Quite how you think that's anything to do with the heatshrink is beyond me.
find another thread to harass ok?

Noto
Posts: 4
Joined: Sat Dec 19, 2009 4:44 pm
Location: Over there

Post by Noto » Sat Dec 19, 2009 5:00 pm

Dear colm,

I've been reading your threads/posts for a while now and I've decided to register to deliever this message.

You speak/type utter, UTTER, pure, unadulterated, weapons grade bollocks. You should chop off your fingers and remove your larynx now to prevent any one else from hearing or seeing your inane drivel.

What's even more worrying is that you actually, genuinely believe the rubbish that you spout and quite frankly I find it amazing that no one else on here other than Monkeh16 has had the balls to speak out against it and even then you simply dismiss what he has to say because obviously you cannot be wrong for you computer knowledge will be written down in years to come and will replace the Bible, Quran, Masoretic Text, Tanakh and you will be revered as God almighty himself.

Some examples of your deluded ramblings include how high micron CPUs somehow mess up photographs. That was my personal favourite until you came out with how your fancy heatshrink somehow managed to increase your CPU voltage and made your fans spin faster. While I don't deny that these events might have taken place my Lord but I find it nigh on impossible to believe that a few pieces of rubber that shrinks when heated can have any effect on the electrical properties of an unrelated component.

In closing Oh Great One, I highly recommend that you put your computer back together, put it down and never, ever open it again nor should you ever type down any of the ideas you get in you head unless you are waging a secret war on everybody on the Internet starting with SPCR where you type mindless crap that causes people's brains to melt and slowly dribble out of their ears until you've finally conquered the Earth.

colm
Posts: 409
Joined: Tue Jan 31, 2006 8:22 am
Location: maine

Burn in update

Post by colm » Sun Dec 20, 2009 1:59 pm

As it turns out my last board wasn't clocking as claimed. I went nto the bios and attempted "burn-in mode" ran an extended memory test and a standard one, vista version. This means no outside software. No errors found, yet in the logs, it claimed it could not load two drivers on system boot. the funnypart of this is...the drivers were not installed, nor the programs associated!

Seems something got sticky. So I rebooted, and now no errors.
Ready to get back to coding/encoding a bit more confident.

Processors Information
------------------------------------------------------------------------------------

Processor 1 (ID = 0)
Number of cores 1 (max 1)
Number of threads 2 (max 2)
Name Intel Pentium 4
Codename Prescott
Specification Intel(R) Pentium(R) 4 CPU 3.40GHz
Package Socket 478 mPGA (platform ID = 2h)
CPUID F.3.4
Extended CPUID F.3
Core Stepping D0
Technology 90 nm
Core Speed 3536.7 MHz (17.0 x 208.0 MHz)
Rated Bus speed 832.2 MHz
Stock frequency 3400 MHz
Instructions sets MMX, SSE, SSE2, SSE3
L1 Data cache 16 KBytes, 8-way set associative, 64-byte line size
Trace cache 12 Kuops, 8-way set associative
L2 cache 1024 KBytes, 8-way set associative, 64-byte line size
FID/VID Control no
Features
cpu-id.

I can tell now the system is a bit more real than the last setup. Aside from being familiar with an all intel retail board (no branding), there is a lower layer that respnds to all commands. such as "balance power savings". I set the cpu minimum to 75%, and it is obvious. Big relief. Alot of setups d not even respond to these power saving things, this board was a bit too zealous, and that is good.

and some other notes related, the case uretjhaned, power supply made safe for humanity and pets and every other soul/flesh carrying living thing, and the heat shrink tubing (the biggest surprise) all helped it.

Next post will hopefully be a raid0 setup to talk about, adding two w3200aaks to the 82801ER chip, a couple of weeks.
for now, going back to default, out of the burn in mode, and actually using integrity.[/quote]

Post Reply