Noncompliance Post New Topic  New Poll  Post A Reply
my profile | directory login | search | faq | forum home

  next oldest topic   next newest topic
» Noncompliance » Computer Forums » Computer Hardware Discussion » Computer Upgrades

 - UBBFriend: Email this page to someone!    
Author Topic: Computer Upgrades
Klaus
Administrator
Member # 66

Icon 14 posted      Profile for Klaus   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
Well after putting up with my computer case being open and a fan blowing directly on the CPU/video card for 2 years, I upgraded last night. I stopped by one of the computer stores on University ave last night on my way home and picked up an Athlon Thunderbird 1.3g and ASUS motherboard. I took a bit of a hit on the CPU price @ $149 (currently $125 on the internet) but at least I saved shipping and had it right away. After looking for a Win98se OEM cd key for 2 hours (finally called someone from work) – it ran windows setup in about 7 minutes. The formally virus riddled system is now up and running and boots into windows in about 10 seconds.

I guess my purpose for writing this is to encourage anyone who has been thinking about upgrading their system – that now might be the time. Looks like CPUs are dropping in price every week and memory is about $24 for 256 PC133 now.

Check this link for the best prices.
www.sharkyextreme.com/hardware/weekly_cpu/

Klaus

[This message has been edited by Klaus (edited 08-07-2001).]

[This message has been edited by Klaus (edited 08-07-2001).]


Posts: 5484 | From: St. Paul, Mn | Registered: Apr 2002  |  IP: Logged | Report this post to a Moderator
Hauserdaddy
Administrator
Member # 50

Icon 6 posted      Profile for Hauserdaddy     Send New Private Message       Edit/Delete Post   Reply With Quote 
Heya klaus, i am going to be buying a 1.3 gig athlon soon also. Was wondering what one you got? 266fsb or 200fsb? I have heard good and bad things about both. You also got a mobo that has the sdram slots too then? i am not sure if i want to get the 2100 ram motherboard or not. How does it run for you and what board did you end up with?

Thanks for the info...

Hauser


Posts: 2220 | From: Prescott, WI, USA | Registered: Apr 2001  |  IP: Logged | Report this post to a Moderator
Klaus
Administrator
Member # 66

Icon 6 posted      Profile for Klaus   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
I believe I have the 266fsb version. I am running it at 133x10 or 266x5 depending on where you look in the BIOS. I bought the ASUS A7V133/WOA board, and I can only run PC133/100 memory in it not the DDR. They didn’t carry the board that supports both, so I guess I am stuck with PC133 for awhile (but that’s ok I have a lot of it and the board supports 1.5g of memory)

So far it has run perfectly – no lockups or crashes even with power management on. I am running SETI all day today, so hopefully it will still be running when I get home. I might try to overclock it to 1.5g tonight – supposedly it shouldn’t be an issue. All the Thunderbirds over 1.2g are not multiplier locked so overclocking is easier now (fsb can be increased or mult. can be increased) I will keep on eye on how hot the chip gets at different speeds until I find one I like.

A couple other things I like about the motherboard are: there are 2 additional IDE connectors on the motherboard for a total of 4 (all ATA100) and it comes with 4 USB ports. And no more ISA slots! I would stick with either an ASUS or ABIT board stay away from MSI (makers of my previous setups glitchy motherboard) Let me know if you have any other questions.

Klaus


Posts: 5484 | From: St. Paul, Mn | Registered: Apr 2002  |  IP: Logged | Report this post to a Moderator
RockLobster
Rotor Head
Member # 45

Icon 8 posted      Profile for RockLobster     Send New Private Message       Edit/Delete Post   Reply With Quote 
1.3 isnt fast enough for you! Jesus. You might avoid more opencase/hangingfan/lock-up/autoshutof/overheat problems by leaving it at the speed it was manufactured for. When us engineers design something and label it with a cetain performance. It is because we did some high dollar math and included a safety factor to ensure performance/reliablity. Circumventing this only causes people to bitch about why some product they bought dosnt work. Well....when you use something beyond its design limits don't blame us when it F$^(*^%CKIN' blows up!!!

Just my two cent's.

(And, I am only kidding around)


Posts: 2331 | From: Rosemount | Registered: Mar 2001  |  IP: Logged | Report this post to a Moderator
Klaus
Administrator
Member # 66

Icon 6 posted      Profile for Klaus   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
Well, maybe in the HVAC engineering community specifications are set exactly as they are but that’s not always the case in the CPU manufacturing industry. Many times chips come out in batches and test higher then they are later marked as. For example, when I purchased my Athlon 500 – AMD was making only 600-750 chips but there was a large demand for 500’s since they were so cheap. Do you think AMD went back and whipped up some 500’s? Nope, they just labeled some 600 and 650 chips as 500, locked the multiplier at 5 (5 x 100mhz) and shipped them. By simply snapping the case off my Athlon and looking at the cpu I confirmed that it was indeed a 650 chip. A small device attached to the chip allowed me to unlock the multiplier and overclock to 650 or more. My problem was never the CPU itself, it ran cool – but the POS MSI motherboard that was rushed to the market, which MSI later tried to pretend like they never made by removing all traces of it from their web site.

These days AMD doesn’t multiplier lock their 1.2g and above Thunderbirds some think to allow them to be easily overclocked and therefore create more demand for them. It’s always a crap shoot when it comes to overclocking anyways – my 1.3 might hit 1.6 or it might not go over 1.4.

Klaus


Posts: 5484 | From: St. Paul, Mn | Registered: Apr 2002  |  IP: Logged | Report this post to a Moderator
RockLobster
Rotor Head
Member # 45

Icon 8 posted      Profile for RockLobster     Send New Private Message       Edit/Delete Post   Reply With Quote 
It may be true that they were labeled differently but i didnt only study HVAC i also studied Heat Transfer In electronic Equipment. To be sure there were more reasons behind the scenes that they were labeled 500 other than the fact they had the chip laying around. Like something to do with heat transfer. Because simple TD will tell you that the mother board would have little to do with a chip overheating physically. But alot to do with the settings by the user or attaching non factory approved "snap on" devices to allow overclocking of a chip that was manufactured in a Slot A cassing built for a 500 and included a safety that was built in to prevent overclocking and engineer calculated malfunctions. Im not saying that nothing that is not factory approved will ever work its just that there is a reason that its not factory approved. Just like modifiing a truck for off-road use there are certain risks in doing so and it is much more difficult to know the risks when it comes to computer hardware since it is not apparantly evident. You understood the nature of the risk and chose to mess with it anyway. I doubt that the chip would have overheated if you left it at 500, but maybe.

[This message has been edited by Chadwick (edited 08-07-2001).]


Posts: 2331 | From: Rosemount | Registered: Mar 2001  |  IP: Logged | Report this post to a Moderator
Klaus
Administrator
Member # 66

Icon 4 posted      Profile for Klaus   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
Like I said the chip never ran hot, but the case temperature was on the high side and I always got video related lock ups. Correctly clocking my 500@650 resulted in no change in temperature. However, overclocking to 700 I had to up the core voltage a bit and there was an increase in heat but it was still within specs set by AMD. I guess the conventional wisdom on the slot A Athlon was that they ran hot and a good case cooling solution was needed even without overclocking. My solution was to have my case open and a fan pointed at the cpu and video card. After 2 years of that I started to get annoyed and tried to run with the case shut – worked for awhile but then the old issues came back and I had to open my case up again.

Two years ago when I picked up my Athlon 500 (with a 650 core) it was a well known fact amongst overclockers that at that time AMD’s batches were turning out so good (all the chips were testing at 650+) that they didn’t have enough 500’s to meet demand. So they labeled the 650 as 500 – this had nothing to do with engineering and everything to do with supply & demand (something I know a thing or two about After awhile the batches were identified and web sites started to offer the “650” (that was actually labeled a 500) and the “snap on device” to overclock it for less $$ then the real 650. This of course only last for several months until the 500 was phased out.


Posts: 5484 | From: St. Paul, Mn | Registered: Apr 2002  |  IP: Logged | Report this post to a Moderator
Hauserdaddy
Administrator
Member # 50

Icon 6 posted      Profile for Hauserdaddy     Send New Private Message       Edit/Delete Post   Reply With Quote 
umm i have read articles and seen reviews of some real good overclockers getting the t-bird 1.3's up to 2gigs. Not that any of us have the gear for that but it can be done.
The chip can handle the speed its everything else that gets messed up. Also read something about using a pencil to complete some circuit or core something on the chip for overclocking them. Might want to look into that. Also, i think i am going to get a ddr mobo for the cpu i get. That way we can actually compare the two and see which one is actually a better buy when it comes to stability and such. i think i saw a real nicely reviewed one for about 150 bucks heh.
Are you using the retail fan on the cpu or did you buy a different one? I am wondering because i can get the oem chip for a lot cheaper and them just buy a nice aftermarket cooler for it.
Thanks for the info!

Hauser


Posts: 2220 | From: Prescott, WI, USA | Registered: Apr 2001  |  IP: Logged | Report this post to a Moderator
Klaus
Administrator
Member # 66

Icon 6 posted      Profile for Klaus   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
Chad is just mad because Intel chips don't overclock so good j/k Chad.

I bought an OEM chip and an aftermarket cpu cooler (don't remember the name) seems to run cool so far.

From what I understand you shouldnt need to perform the "pencil trick" to overclock any Thunderbird over 1.2 (they aren't mul. locked) Basically what the trick does is connect a group of circuits on the board by drawing on the board inbetween them to make a connection - I guess this allows the muliplier to be unlocked.

Again, make sure to get a ASUS or ABIT board that allows for overclocking. I know my board is the first that allows 1mhz increments in fsb. My old board you only had like 3 options - 100,112,133.

Goodluck


Posts: 5484 | From: St. Paul, Mn | Registered: Apr 2002  |  IP: Logged | Report this post to a Moderator
Mikey
Po Po
Member # 42

Icon 2 posted      Profile for Mikey     Send New Private Message       Edit/Delete Post   Reply With Quote 
I think I have to agree with Chad here. Even though that does hurt me a little. Not because I know anything about heat transfer or pencil snapping or chippies jumpers or crazy shit like that. I look at things on a fundemental basis. What really is the difference between 1.2 gig and 1.5 gig? I guess I'm not a computer whiz but my guess would be not that much. I just think that a person is setting themselves up for some trouble for very little gain by pencil jerking and whizbanging. Anyway, just my 2 cents worth.
Posts: 486 | From: Eagan, MN | Registered: Oct 2000  |  IP: Logged | Report this post to a Moderator
Jomamason
Junior Member
Member # 14

Icon 12 posted      Profile for Jomamason     Send New Private Message       Edit/Delete Post   Reply With Quote 
pencil jerking and whizbanging???? Doesn't sound like computer slang, what are you guys doing with your PCs. lol


Posts: 67 | Registered: Jan 2001  |  IP: Logged | Report this post to a Moderator
RockLobster
Rotor Head
Member # 45

Icon 4 posted      Profile for RockLobster     Send New Private Message       Edit/Delete Post   Reply With Quote 
I guess thats the real issue when it comes to this stuff the risks and troubles that come with overclocking seem to so much outweight the benefits it just dosnt make much sence uless you get some kind of kick out of "pencil jerking" your computer. Im not and "overclocker and dont pretend to be one by any means or know all the little tricks and magic. The simple fact is that the speed diffences you notice are between the different generations of chips and their associates system architecture changes like ram speed and front side buss. You won't notice a difference if you change chips in the same architecture for a faster one or overclock one. The only time you might notice a difference is when running processor intensive applicaitons like AutoCAD 3D or PROEngineer. I guess my feeling is that i would prefer to spend the extra money for the faster chip like 1.5 vs 1.3 and know that it all is going to work or i can return it for a replacement.
Posts: 2331 | From: Rosemount | Registered: Mar 2001  |  IP: Logged | Report this post to a Moderator
Klaus
Administrator
Member # 66

Icon 6 posted      Profile for Klaus   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
quote:
Originally posted by mikey:
I think I have to agree with Chad here. Even though that does hurt me a little. Not because I know anything about heat transfer or pencil snapping or chippies jumpers or crazy shit like that. I look at things on a fundemental basis. What really is the difference between 1.2 gig and 1.5 gig? I guess I'm not a computer whiz but my guess would be not that much. I just think that a person is setting themselves up for some trouble for very little gain by pencil jerking and whizbanging. Anyway, just my 2 cents worth.

I agree with you Mike that a 1.3g @ 1.5g isn’t really a big increase (7.69%) and I would just be seeing if my chip would run at a higher speed, but I am not planning on keeping it overclocked. Now in your case – you have a 800 Thunderbird, right? What if by drawing a couple lines on the chip with a pencil you could run it at 1.0g or 1.2g? A 25% or 50% increase might be worth it then, right? I know you would rather just spend the $100 and get a 1.0g or 1.2g and that cool J


Posts: 5484 | From: St. Paul, Mn | Registered: Apr 2002  |  IP: Logged | Report this post to a Moderator
Klaus
Administrator
Member # 66

Icon 6 posted      Profile for Klaus   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
I am running a PIII 600(100fsb x 6) at work and I do a lot of Photoshop and some AutoCad with it. When I bumped up the fsb to 133 making it an 800 I did notice a difference in performance.
Posts: 5484 | From: St. Paul, Mn | Registered: Apr 2002  |  IP: Logged | Report this post to a Moderator
RockLobster
Rotor Head
Member # 45

Icon 6 posted      Profile for RockLobster     Send New Private Message       Edit/Delete Post   Reply With Quote 
It's all a big farce. ;,
Posts: 2331 | From: Rosemount | Registered: Mar 2001  |  IP: Logged | Report this post to a Moderator
RockLobster
Rotor Head
Member # 45

Icon 13 posted      Profile for RockLobster     Send New Private Message       Edit/Delete Post   Reply With Quote 
You do understand Scott that you just dont get a straight percentage speed increase. That your speed and performance depend on alot of other factors which can easily muddle a 25% speed increase in an identical system architecture.
Posts: 2331 | From: Rosemount | Registered: Mar 2001  |  IP: Logged | Report this post to a Moderator
Klaus
Administrator
Member # 66

Icon 6 posted      Profile for Klaus   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
quote:
Originally posted by Chadwick:
You do understand Scott that you just dont get a straight percentage speed increase. That your speed and performance depend on alot of other factors which can easily muddle a 25% speed increase in an identical system architecture.

Did I ever say that the percentage increase of Mhz equated to the percentage increase in “performance”? I just don’t agree with your statement that Mhz within the same architecture are all the same (ie. PIII800 = PIII900) - that’s just silly.


Posts: 5484 | From: St. Paul, Mn | Registered: Apr 2002  |  IP: Logged | Report this post to a Moderator
RockLobster
Rotor Head
Member # 45

Icon 6 posted      Profile for RockLobster     Send New Private Message       Edit/Delete Post   Reply With Quote 
True, anyway this topic is startin to get boring.

What is the name of that place that you found that cheap memory at of university? The micron stuff not the crappy PNY stuff that G Nanosystems and best buy sell.


Posts: 2331 | From: Rosemount | Registered: Mar 2001  |  IP: Logged | Report this post to a Moderator
Klaus
Administrator
Member # 66

Icon 6 posted      Profile for Klaus   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
Thank you for ending this I appreciate it (I need to get some work done), I will buy you a beer next time I see you .

Tran Micro Computer – 612-379-2572 – 2720 University Ave
General NanoSystems, Inc. - 612-331-3690 – 3014 University Ave

I think Tran Micro had the Micro memory, but I can’t remember.


Posts: 5484 | From: St. Paul, Mn | Registered: Apr 2002  |  IP: Logged | Report this post to a Moderator
Cramer
Poser
Member # 5

Icon 8 posted      Profile for Cramer     Send New Private Message       Edit/Delete Post   Reply With Quote 
ALL COMPUTERS SUCK ASS!!!

EVERY TIME!!!


Posts: 771 | From: Farmington | Registered: Oct 2000  |  IP: Logged | Report this post to a Moderator
Cramer
Poser
Member # 5

Icon 10 posted      Profile for Cramer     Send New Private Message       Edit/Delete Post   Reply With Quote 
they really do!!!
Posts: 771 | From: Farmington | Registered: Oct 2000  |  IP: Logged | Report this post to a Moderator
Cramer
Poser
Member # 5

Icon 8 posted      Profile for Cramer     Send New Private Message       Edit/Delete Post   Reply With Quote 
im not kidding!!!
Posts: 771 | From: Farmington | Registered: Oct 2000  |  IP: Logged | Report this post to a Moderator
BoondockSaint
Moderator
Member # 67

Icon 11 posted      Profile for BoondockSaint     Send New Private Message       Edit/Delete Post   Reply With Quote 
Just when you thought you had a fast machine...the GeForce 4 comes out on Feb. 19th. Hooray! $400 will be spent as soon as I HAVE it.

Good god man, 128 MB of DDR RAM...

[This message has been edited by BoondockSaint (edited 02-11-2002).]


Posts: 1845 | From: Chaska | Registered: Apr 2002  |  IP: Logged | Report this post to a Moderator
BoondockSaint
Moderator
Member # 67

Icon 2 posted      Profile for BoondockSaint     Send New Private Message       Edit/Delete Post   Reply With Quote 
Correction: It looks like a 1st week in March release...
Posts: 1845 | From: Chaska | Registered: Apr 2002  |  IP: Logged | Report this post to a Moderator
Klaus
Administrator
Member # 66

Icon 13 posted      Profile for Klaus   Author's Homepage     Send New Private Message       Edit/Delete Post   Reply With Quote 
I don't know Eric, it dosen't look like the GF4 is really that huge of a step forward in technology(for the price)....


Nvidia's GeForce4 Hits and Misses
By Vince Freeman

The 4 Ti Is a Clear Advance; the 4 MX Is False Advertising
The 3D graphics market got a lot hotter today, as Nvidia announced new products spanning all market or price segments. The GeForce4 graphics processor line represents a huge shift for the company -- in the past, each new Nvidia product offered a quantum leap forward in either processing power or features, along with a new core design. The high-performance GeForce4 Ti and value-priced GeForce4 MX are quite different in that respect, being more of enhancements to existing GeForce3 and GeForce2 technologies, respectively.

At the high end, there's the true GeForce4 chip, which is a higher-clocked, core-enhanced version of Nvidia's former flagship GeForce3 Ti 500. There are some new bells and whistles, such as an extra vertex shader, new antialiasing engine, and improved, 128-bit DDR memory controller, but basically the 4 Ti still looks a lot like a faster GeForce3.

The 4 comes in three flavors -- the ultra-high-end GeForce4 Ti 4600, the high-end GeForce4 Ti 4400, and the midrange GeForce4 Ti 4200. The last offers performance at about the level of a GeForce3 Ti 500, while the upper two are even faster.

While ATI is already sniping about "GeForce3.5," the new name is at least partly deserved for the true GeForce4 cards. The new boards' prices will be welcome news, with the GeForce4 Ti 4200 undercutting GeForce3 Ti 500 prices, while the higher-speed models stay competitive with Nvidia's previous performance-card pricing. So bleeding-edge game crazies get higher performance at the same price, while the rest of us get a new midrange board that's likely to be popular with a far larger market segment. It may be rather like Intel hypothetically introducing the Northwood as the Pentium 5, but as long as price/performance improves, the label is largely inconsequential.

There's a reason I'm referring to the 4 Ti as "the true GeForce4," however: things get a lot more muddy when considering the new GeForce4 MX. Whether or not you're cool with the new GeForce4 naming scheme, the GeForce4 MX -- available in MX 460, MX 440, and MX420 versions -- is actually more of a GeForce2 MX derivative with some enhancements thrown in for good measure.

For the uninitiated, the GeForce2 MX was a stripped-down version of the GeForce2, with only half of the processing pipelines found in the original GeForce2 core. This helped Nvidia achieve a lower price and made the GeForce2 MX a popular (though lower-performing) choice with system OEMs, both on separate graphics cards and in Nvidia's nForce integrated chipset.

You Are the Weakest Link, Goodbye
Unfortunately, the new MX is a GeForce4 in name only. The GeForce4 MX has none of the nifty DirectX 8/8.1 features supported by in Nvidia's nfiniteFX engine, and even the former midrange GeForce3 Ti 200 offering is a far more advanced and feature-rich chip. There have been a few enhancements such as a faster memory controller and hardware DVD playback, but these exist outside the core architecture. Again, even though the GeForce2 MX was a hobbled product for the value market, it was based on a GeForce2 core. By contrast, I'd be hard pressed to find a reason to call Nvidia's new value entry even a GeForce3 MX.

This move might even indicate a plateau effect in 3D technology. When Nvidia started its GeForce3 marketing push, DirectX 8 and nfiniteFX support were placed on a pedestal and their importance to future 3D development was proclaimed. Now with the GeForce4 MX not even including these basic features, what kind of product support can GeForce3 owners expect?

Frankly, I think the GeForce4 MX announcement immediately tarnishes the GeForce4 name and makes Nvidia's numbering system nearly useless. I'm quite familiar with the market, but can still find vendors' myriad names and numbers confusing. Imagine, then, how it'll look to the average consumer, who buys a new "High-Performance GeForce4" desktop for Junior only to find that its GeForce4 MX 420 SDR video is actually slower than the old GeForce2 Ultra?

These are very real risks for Nvidia -- except for streamlining its varied product line (even mobile chips) under a single name, what could the company hope to gain by letting the hobbling MX version share the prestigious GeForce4 label with the real high-performance products? Well, the company took a lot of flak for including "only" GeForce2 MX video in its nForce chipset; maybe it'll sound better to advertise the next nForce as incorporating GeForce 4 MX graphics. Also, to be fair, Nvidia might feel that the general computing public doesn't want or need the immense power of a true GeForce4 or even GeForce3, but acknowledge that its economy TNT2 and Vanta products are past due for retirement.

Whatever the reasons, the DirectX-7-generation GeForce4 MX will spell confusion in the marketplace, and I wouldn't imagine Nvidia's competitors are happy with this slight of hand. Imagine the ads we'll see touting GeForce4 systems at prices well below those of ATI Radeon-equipped PCs. There'll likely be a lowball GeForce4 MX inside, but it's an easy bet which system will get more attention.

So I'm tossing some laurels to Nvidia for leading the 3D performance race with the impressive GeForce4 Ti, but throwing darts at the deceitfully labeled GeForce4 MX -- it's not that the 4 MX is a poor performer for its market, but it clearly flaunts a name well above its station and hopes to beat its competitors with branding alone. Time will tell if the ploy works, but consumers like an ordered world with consistent numbering. Marketing a pumped-up GeForce2 MX as a GeForce4 seems to be stepping well over the line, and might give ATI some extra market share based on consumer goodwill alone.


Posts: 5484 | From: St. Paul, Mn | Registered: Apr 2002  |  IP: Logged | Report this post to a Moderator
Cramer
Poser
Member # 5

Icon 3 posted      Profile for Cramer     Send New Private Message       Edit/Delete Post   Reply With Quote 
" megahertz this, gigabytes that, you just smile and nodd, knowing you havnt got a clue "

god-damn I find that radio spot funny.....
it is sooooooo me!!!

[This message has been edited by cramer (edited 02-19-2002).]


Posts: 771 | From: Farmington | Registered: Oct 2000  |  IP: Logged | Report this post to a Moderator
   

Quick Reply
Message:

HTML is enabled.
UBB Code™ is enabled.

Instant Graemlins
   


Post New Topic  New Poll  Post A Reply Close Topic   Feature Topic   Move Topic   Delete Topic next oldest topic   next newest topic
 - Printer-friendly view of this topic
Hop To:


Contact Us | Noncompliance.com

Noncompliance Copyright 2005

Powered by Infopop Corporation
UBB.classic™ 6.7.2