Trashbyteswrote on 2024-08-26, 07:57:Got this nice clean ASRock 939Dual Sata2 for a good price, seems like the perfect board for the 4800+ or an Opteron 180, they ar […] Show full quote
Got this nice clean ASRock 939Dual Sata2 for a good price, seems like the perfect board for the 4800+ or an Opteron 180, they are essentially the same CPU just depends on which one I can dig out of the CPU box first.
Comes with everything except the manual, also doesnt appear to have any bad caps but will get a good testing once its here.
I sold dozens of those back in the day, quite literally every single one I sold or used in my shop had capacitors that failed after a few years of usage.
Updated my previous message. Its actually either December 1998 or very early 1999. https://www.cnet.com/tech/tech-industry/amd-c … ntel-in-retail/ article has wrong date attribution. Listed Retail channel best sellers, aka mum gets lost in a supermarket and spends too much on bad computer after being bombarded with slang by Computer salesperson.
1 HP Pavilion 6330 - AMD K6-2 300 MHz, 48MB memory (how?), SiS 5598 integrated "graphics". $800 + $300 15" M50 monitor
2 Compaq Presario 5030 - Pentium II 300 MHz, 64MB, integrated ATi Rage LT, Zip drive included! 17" MV700 monitor. $1650
3 Compaq Presario 2256 - AMD K6 300 MHz, 48MB, 430VX with integrated 2MB S3 Virge. $900 + ~$260 15" monitor
4 Compaq Presario 5020 - Celeron 300 MHz, 64MB, integrated ATi Rage LT. $1000 + ~$260 15" monitor
5 Apple iMac (233 MHz), 32MB, 6 MB ATI Rage Pro Turbo. $1300
6 IBM Aptiva something K6-2 300 MHz. Who knows how much.
7 Packard Bell Club 30 - Cyrix MII 300 CPU 233 MHz, 32MB, SiS 5598 integrated "graphics", 14" monitor. ~$800
iMac most expensive but surprisingly best choice from this big store dreg. No 3D games on any of them tho 😀 all for homework, senior email service and taxes.
Trashbyteswrote on 2024-08-26, 07:57:Got this nice clean ASRock 939Dual Sata2 for a good price, seems like the perfect board for the 4800+ or an Opteron 180, they ar […] Show full quote
Got this nice clean ASRock 939Dual Sata2 for a good price, seems like the perfect board for the 4800+ or an Opteron 180, they are essentially the same CPU just depends on which one I can dig out of the CPU box first.
Comes with everything except the manual, also doesnt appear to have any bad caps but will get a good testing once its here.
I sold dozens of those back in the day, quite literally every single one I sold or used in my shop had capacitors that failed after a few years of usage.
Otherwise I like that board a lot.
I have two versions of its 775 cousin too, but yes their caps are pretty terrible. This one "Looks" ok but I will keep an eye on it and have some replacement caps on hand in case it pops, the Opteron 180 may push them enough to pop any bad ones.
- ASUS P5GC-MX
- KME CX-6058 case
- Rexpower PSU (also KME OEM)
- Geforce 8500GT 512MB that was artefacting
- 250GB IDE WD Caviar SE
- 1GB DDR2
- Samsung SH-S203 DVDRW (SATA)
Pic from when I was cleaning it up.
deas for a build inside it? I'm thinking of the following:
- 865 based 775 build
- ASRock P4Dual-915GL w/ Radeon X1950 XTX and Prescott P4
- mobile Barton build with either SL-75FRN2-RL or NF7-S v2.0
- (reserved space for special Socket 462 build)
- ASUS P5GC-MX
- KME CX-6058 case
- Rexpower PSU (also KME OEM)
- Geforce 8500GT 512MB that was artefacting
- 250GB IDE WD Caviar SE
- 1GB DDR2
- Samsung SH-S203 DVDRW (SATA)
Pic from when I was cleaning it up.
deas for a build inside it? I'm thinking of the following:
- 865 based 775 build
- ASRock P4Dual-915GL w/ Radeon X1950 XTX and Prescott P4
- mobile Barton build with either SL-75FRN2-RL or NF7-S v2.0
- (reserved space for special Socket 462 build)
Thats wild i just picked up the exact same case, from the same estate sale I got the Compaq from. Except it has a crazy surge protector strip built in. I'll have to grab a pic of it. The lan party handle is the best thing!
Who decides what truth is, and what is their objective? Today’s falseness can reappear as tomorrow’s truth.
Yeah, this is my 2nd one.
The other one I have (IIRC the 6059) is white/khaki, and it houses a DFI Infinity nF4 SLi w/ an Athlon 64 x2 4600+ and a Geforce 7900GTX KingKong Edition.
I wish I had the funds to swap the board for a Lanparty counterpart though.
"Enter at your own peril, past the bolted door..."
Main PC: i5 3470, GB B75M-D3H, 16GB RAM, 2x1TB
98SE : P3 650, Soyo SY-6BA+IV, 384MB RAM, 80GB
Trashbyteswrote on 2024-08-26, 07:57:Got this nice clean ASRock 939Dual Sata2 for a good price, seems like the perfect board for the 4800+ or an Opteron 180, they ar […] Show full quote
Got this nice clean ASRock 939Dual Sata2 for a good price, seems like the perfect board for the 4800+ or an Opteron 180, they are essentially the same CPU just depends on which one I can dig out of the CPU box first.
Comes with everything except the manual, also doesnt appear to have any bad caps but will get a good testing once its here.
I sold dozens of those back in the day, quite literally every single one I sold or used in my shop had capacitors that failed after a few years of usage.
Otherwise I like that board a lot.
It can be picky about graphics cards.
I use one as main test system for graphics card i buy but it's not the silver bullet. I tested a lot of agp and PCI graphics cards that just give black screen on this system. So i often have a second motherboard at hand to confirm.
Also, a few days ago, when i talked about my opteron 180 that stopped working. This was after using it in this board (For a "GPU June" video). Not sure if it'the board fault or ESD when i manipulated it.
aclwrote on 2024-08-27, 08:11:It can be picky about graphics cards.
I use one as main test system for graphics card i buy but it's not the silver bullet. I te […] Show full quote
Trashbyteswrote on 2024-08-26, 07:57:Got this nice clean ASRock 939Dual Sata2 for a good price, seems like the perfect board for the 4800+ or an Opteron 180, they ar […] Show full quote
Got this nice clean ASRock 939Dual Sata2 for a good price, seems like the perfect board for the 4800+ or an Opteron 180, they are essentially the same CPU just depends on which one I can dig out of the CPU box first.
Comes with everything except the manual, also doesnt appear to have any bad caps but will get a good testing once its here.
I sold dozens of those back in the day, quite literally every single one I sold or used in my shop had capacitors that failed after a few years of usage.
Otherwise I like that board a lot.
It can be picky about graphics cards.
I use one as main test system for graphics card i buy but it's not the silver bullet. I tested a lot of agp and PCI graphics cards that just give black screen on this system. So i often have a second motherboard at hand to confirm.
Also, a few days ago, when i talked about my opteron 180 that stopped working. This was after using it in this board (For a "GPU June" video). Not sure if it'the board fault or ESD when i manipulated it.
Agp motherboards in general seem to be randomly picky about the graphics cards you put in them. I have come to find you need to test all video cards in at LEAST two motherboards with different chipsets before declaring an agp gpu is dead.
To many times I have dropped in a card only to think its dead, then the next motherboard works perfectly. I even had one card reject Nforce 3 & Via K8T800 Pro chipsets, only to run perfectly on another with an intel chipset.
Who decides what truth is, and what is their objective? Today’s falseness can reappear as tomorrow’s truth.
aclwrote on 2024-08-27, 08:11:It can be picky about graphics cards.
I use one as main test system for graphics card i buy but it's not the silver bullet. I te […] Show full quote
I sold dozens of those back in the day, quite literally every single one I sold or used in my shop had capacitors that failed after a few years of usage.
Otherwise I like that board a lot.
It can be picky about graphics cards.
I use one as main test system for graphics card i buy but it's not the silver bullet. I tested a lot of agp and PCI graphics cards that just give black screen on this system. So i often have a second motherboard at hand to confirm.
Also, a few days ago, when i talked about my opteron 180 that stopped working. This was after using it in this board (For a "GPU June" video). Not sure if it'the board fault or ESD when i manipulated it.
Agp motherboards in general seem to be randomly picky about the graphics cards you put in them. I have come to find you need to test all video cards in at LEAST two motherboards with different chipsets before declaring an agp gpu is dead.
To many times I have dropped in a card only to think its dead, then the next motherboard works perfectly. I even had one card reject Nforce 3 & Via K8T800 Pro chipsets, only to run perfectly on another with an intel chipset.
I have seen lots of picky AGP cards\boards too.
I think part of the time these quirks can be blamed on the physical AGP interface itself. I have seen situations where a card can be latched in place and screwed in and yet still have some play within the socket - sometimes making the system not POST unless the card is pressed on a bit. It can be even worse with older cards and boards that don't have latches.
If you compare the contacts on an AGP card vs ISA, PCI and PCI-Express, you'll see that the shape of the pins is very different on AGP and they are staggered. I don't know if this is why some combinations of cards + boards can be so picky, but it could be why we don't see connectors like this on consumer gear these days (there may be some, but I can't think of any). It seems like that would have been a pretty huge oversight in the design, but they may have had other good reasons to make the connector this way. Maybe they did this to manage the force required to insert or remove cards? If half the pins are at different depths then it would require half the force spread out over two rows of pins.
But then, this also makes the card come loose more easily too. Does anyone remember opening up PCs back in the late 90s and early 2000s and finding hot glue on the AGP connector to make sure the video card didn't lose connection in shipping? Good times. 🤣
Going along with all of this, if an AGP card just doesn't seem to "agree" with a certain board, you do not have any other slots to try it in, so if there's anything even remotely weird going on with the slot or the pins within the slot you have no way to troubleshoot except to try the card in a different board. If it works there, then it can give the appearance that the board or chipset is to blame. I have gone back and forth with cards in different boards only to have it work in the original board eventually.
Yes, a mostly good condition G4 Cube. Only real problems are the slightly sticky cabling and the dry-rotting speakers. Both are kind of expected issues, since it's a 24 year old computer.
It seems like that would have been a pretty huge oversight in the design, but they may have had other good reasons to make the connector this way. Maybe they did this to manage the force required to insert or remove cards? If half the pins are at different depths then it would require half the force spread out over two rows of pins.
Given the number of pins, if AGP card edges weren't staggered it would end up being like VLB all over again. No-one wanted the "thousand pin apocalypse".
It seems like that would have been a pretty huge oversight in the design, but they may have had other good reasons to make the connector this way. Maybe they did this to manage the force required to insert or remove cards? If half the pins are at different depths then it would require half the force spread out over two rows of pins.
Given the number of pins, if AGP card edges weren't staggered it would end up being like VLB all over again. No-one wanted the "thousand pin apocalypse".
Normal AGP only has around 16 more pins than PCI, so I don't think it was purely a connector size issue. AGP Pro is, of course, much larger and has more pins, but that came several years later and could have just as well never been created (aux power connectors on cards made AGP Pro irrelevant). Perhaps they anticipated AGP getting larger over time and were expecting PC motherboards to shrink over that same time period, necessitating a more dense slot, but AGP 2x, 4x and 8x all work in the same size slot, so I don't know how much that really affected things.
PCI-Express x16 slots have 32 more pins than AGP and are longer than AGP, but the designers seemingly didn't feel the need to stagger the pin depth to save space.
Anyway, it is surprising to me just how little information exists about the AGP standard and why it was made the way it was. Or, I should say, my Google search attempts are mostly coming up with barely-related results and nothing very technical. I would be curious to know exactly why it was made the way it was.
The only standard PC interfaces I can think of that use staggered pin-depth connectors like these are EISA, Slot 1, Slot A and AGP. I don't think any of these are known for being exceptionally reliable or stable connectors, which may be why PCI-E and other interfaces moved away from staggered pins.
Ozzuneojwrote on 2024-08-28, 03:30:Normal AGP only has around 16 more pins than PCI, so I don't think it was purely a connector size issue. AGP Pro is, of course, […] Show full quote
It seems like that would have been a pretty huge oversight in the design, but they may have had other good reasons to make the connector this way. Maybe they did this to manage the force required to insert or remove cards? If half the pins are at different depths then it would require half the force spread out over two rows of pins.
Given the number of pins, if AGP card edges weren't staggered it would end up being like VLB all over again. No-one wanted the "thousand pin apocalypse".
Normal AGP only has around 16 more pins than PCI, so I don't think it was purely a connector size issue. AGP Pro is, of course, much larger and has more pins, but that came several years later and could have just as well never been created (aux power connectors on cards made AGP Pro irrelevant). Perhaps they anticipated AGP getting larger over time and were expecting PC motherboards to shrink over that same time period, necessitating a more dense slot, but AGP 2x, 4x and 8x all work in the same size slot, so I don't know how much that really affected things.
PCI-Express x16 slots have 32 more pins than AGP and are longer than AGP, but the designers seemingly didn't feel the need to stagger the pin depth to save space.
Anyway, it is surprising to me just how little information exists about the AGP standard and why it was made the way it was. Or, I should say, my Google search attempts are mostly coming up with barely-related results and nothing very technical. I would be curious to know exactly why it was made the way it was.
The only standard PC interfaces I can think of that use staggered pin-depth connectors like these are EISA, Slot 1, Slot A and AGP. I don't think any of these are known for being exceptionally reliable or stable connectors, which may be why PCI-E and other interfaces moved away from staggered pins.
I have a book "Computer buses design and application" W. Buchanan.
With some details about AGP (and other buses) implementation. Unfortunately nothing about the
staggered pins design but there are information about the extra pins over PCI.
This is not a great book... but it contains interesting details and can be found as PDF online (it's from 2000 and probably out of print)
Sample from "Computer buses design and application" W. Buchanan.
File license
Fair use/fair dealing exception
I personally think that the staggered pins design is just more compact. As motherboards became more and more integrated (with sound, usb, disk/raid controllers) physical size could become a problem for manufacturing. And if you want your new standard to be supported by the manufacturers you want to lower the cost and complexity barrier.
I read that it's also why AMD chosed to re-use the Intel Slot1 connector (but rotated) for SlotA...
The connector was already manufactured (for Intel) and available to motherboard manufacturers. Not having to source a different connector lowered the cost of producing an AMD motherboard and made the manufacturers happy.
Yes, a mostly good condition G4 Cube. Only real problems are the slightly sticky cabling and the dry-rotting speakers. Both are kind of expected issues, since it's a 24 year old computer.
