In case you don’t know: Hard drive considerations when building/upgrading a PC

DP
Posted By
Daryl_Pritchard
Jul 21, 2004
Views
1118
Replies
53
Status
Closed
Hello all,

This is likely nothing new to those experienced in rebuilding/upgrading a a computer, but I thought I’d throw it out for those who might benefit from it:

If you have a PC built with current technology, higher performance hard drive(s) and are thinking of adding another older, lower-performance drive that you’ve salvaged from elsewhere, be cautious about mixing old and new on the same IDE channel. Why? Because an IDE channel can only support one transaction speed and this will be restricted to that of the lowest-performance drive installed on that channel. If possible, move the low-performance drive to another channel. If you’re then mixing it with an optical (CD, DVD, etc.) drive, the optical drive will likely introduce a similar performance restriction, but at least your newer hard drive will perform to its full potential. Better yet, forget about salvaging the old drive. Prices what they are, just buy a new hard drive similar to the most current drives in your system.

Case in point…I recently bought a new PC with a single 160GB Western Digital Caviar ATA100 hard drive. While knowing I shouldn’t throw an old ATA66 drive onto the same IDE channel, I did anyway…the other IDE channel is already supporting two optical drives. The old drive was just a temporary measure anyway, for creating some backup images while waiting for the arrival of a new Hitachi 160GB SATA drive.

With the ATA100/ATA66 mix, I ran a test action on a 22MB file that took 55 seconds to complete in PS CS, with the ATA66 drive serving as the PS scratch disk (with 1GB DDRAM, I’m not sure if the scratch disk even got used…I forgot to monitor it). When the new SATA drive arrived, I installed it and removed the ATA66 drive. So, now I had a single ATA100 drive on an unshared IDE channel and a SATA drive similarly on its own. Rerunning the test action, the time improved to 41 seconds!

Although I find it rather hard to believe that throttling down an ATA100 drive to ATA66 performance would result in a 14 second performance penalty on this test action, that at least appears to have been the case. Now, I’ll add too that I didn’t run this test action but once for each drive configuration and that if I had, different numbers would likely have been seen. Regardless, even if multiple test runs had yielded an average time of 7 seconds, that remains quite an improvement.

So, while the performance hit may not be as extreme as I first observed, the point remains that mixing the types of drives on the same IDE channel may have a larger impact than you’d expect and should be avoided if possible.

Regards,

Daryl

Master Retouching Hair

Learn how to rescue details, remove flyaways, add volume, and enhance the definition of hair in any photo. We break down every tool and technique in Photoshop to get picture-perfect hair, every time.

DM
dave_milbut
Jul 21, 2004
Why? Because an IDE channel can only support one transaction speed and this will be restricted to that of the lowest-performance drive installed on that channel

So, while the performance hit may not be as extreme as I first observed, the point remains that mixing the types of drives on the same IDE channel may have a larger impact than you’d expect and should be avoided if possible.

for the same reason, it’s not wise to mix your cd/dvd drives on the same channel as your hard drives.
TI
Thomas_Ireland
Jul 21, 2004
Now I’m confused guys… 🙁

Is this ok:

Primary IDE Master: C Drive (120 GB HD)
Primary IDE Slave: E Drive (Sony DVD +/-R//+/-RW)
Secondary IDE Master D Drive (120 GB HD)
Secondary IDE Slave: F Drive (CD +/-R//+/-RW)

Or is best to:

Primary IDE Master: C Drive (120 GB HD)
Primary IDE Slave: E Drive (120 GB HD)
Secondary IDE Master: D Drive (Sony DVD +/-R//+/-RW)
Secondary IDE Slave: F Drive (CD +/-R//+/-RW)

Or something else? Jeepers! Just when I think I have all the answers, I find out I’m taking the wrong test.

Thanks for I/A for the help.

Tom Ireland
DM
dave_milbut
Jul 21, 2004
tom, it depends on the speed of all those devices. generally cd/dvd’s are slower than hard drives.

Just when I think I have all the answers, I find out I’m taking the wrong test.

heh! it’s easy to get hooked on this place. i started out a couple years ago, peeking at a thread here and a thread there. soon i started answering the stuff i knew about pc’s and os’s, then basic to intermediate PS questions. now 2 1/2 years later, i STILL learn something here almost every day.
DP
Daryl_Pritchard
Jul 21, 2004
Tom,

Your 2nd alternative is the better of the two as far as I know. Even if one of your hard drives was a lesser performing unit than the other, such as the mix of ATA100 and ATA66 hard drives that I first had, it is likely that the performance would still be better with both drives on the same IDE channel than having each drive further throttled down by sharing a channel with an even slower CD or DVD drive.

What I don’t know however, is whether the impact of a CD or DVD drive upon a same-channel hard drive’s performance exists only when a CD/DVD is loaded and being accessed. That is, does an empty, inactive CD/DVD drive still slow down a hard drive? I’m guessing not. Since only the hard drive would be pumping data out onto the IDE interface, I assume full hard drive performance would be delivered. I didn’t originally plan to buy a new drive for this new PC until I thought about how the ATA66 drive would limit the ATA100’s performance. Since drives are cheap, I bought the 160GB SATA for $100 and got rid of the ATA66. Of course I could’ve gone with another ATA100 drive also, but the SATA was on sale at Newegg and I liked the idea of each drive having its own dedicated channel.

This new PC is aching to be my Photoshop system rather than a media center!

Daryl
DM
dave_milbut
Jul 22, 2004
That is, does an empty, inactive CD/DVD drive still slow down a hard drive? I’m guessing not.

I’m not sure, but I think it might. I believe it gets all that data from the drive’s firmware before the os even loads. If it didn’t "Bad Things" could happen if you started to use the cd drive while the hard drive was going at full clip.
DP
Daryl_Pritchard
Jul 22, 2004
Without knowing for sure, what you suggests fits my line of thinking Dave. It definitely seems reasonable that certain hardware operating parameters are established by their firmware whether the hardware is actually actively sending/receiving data or not.
BB
brent_bertram
Jul 22, 2004
Ideally, I think you need to keep each device on its own cable. I’ve been able to do that by adding a dma100 controller as well as a 4 channel RAID controller . I, too, don’t like to risk "mixing and matching" different IDE specs on the same cable.

🙂

Brent
H
Ho
Jul 22, 2004
I’m under the impression that modern IDE controllers can access a drive at its native speed no matter what its sharing the channel with. I researched this once before, but I can’t site any references now. If I have time later today to look into it again, I will.
BB
brent_bertram
Jul 22, 2004
Ho,
Here is a link ( 2002 )supporting my worries, however, it may well be out-of -date information. < http://www.winnetmag.com/Windows/Article/ArticleID/23297/232 97.html> . I’ve been "googling" for a hint that modern chipsets ( or something ) now have solved the problem, but I haven’t found it yet.

[edit] Found it ! "Independent Master/Slave Device Timing: Hard disk controllers on modern systems support running the master and slave device at different speeds, if one supports faster transfer modes than the other. Some systems, however, especially older ones, do not. If you are using two devices with radically different maximum transfer rates, and the chipset doesn’t support independent timing, you will slow down the faster device to the speed of the slower one. "

Article on Storage Review , < http://www.storagereview.com/guide2000/ref/hdd/if/ide/confPe rformance.html> .

🙂

Brent
DP
Daryl_Pritchard
Jul 22, 2004
Brent and Ho,

Thanks for that information. I knew I had seen questions in the past about whether or not optical drives and hard drives could share the same IDE channel without affecting performance, but it always seemed the answers were a mixed lot, with nothing to back them up. I knew that mixing SCSI devices on the same chain is a no-no, and it seemed reasonable that IDE behaved similarly. I guess the kicker is that this was true at one point in time, but not now. So, perhaps my observations with this new PC and the ATA66/ATA100 mix was coincidental? Before I started this thread, I double-checked a few related articles on the net that seemed to back up what I’d seen, but I honestly didn’t pay attention to their date to assess whether they reflected the current state of technology. So, it sounds as though I am wrong in what I sad. I apologize for any confusion this may have created…I thought I was sharing correct information, especially given what I’d observed.

Daryl
BB
brent_bertram
Jul 22, 2004
Daryl,
There’s no apology necessary. PC’s are a such a mixed bag, and we tend to have faith in our own experiences. I’ve been working on PC’s since 1984, and I still have ingrained habits that are out of date. I do think cautious person would always take new claims and new "features" with a dose of skepticism and stay with "tried and true" techniques until you absolutely must embrace the newer technology. I’ve been on the "bleeding edge" and prefer not to be.
Your observations about the new PC may well be valid, but for reasons that we don’t yet understand ( I hate to think about all the things I don’t fully understand <G> ) .

🙂

Brent
DP
Daryl_Pritchard
Jul 22, 2004
Brent,

Thanks…I just hate saying anything that could actually be misleading.

With regard to this new PC, my curiosity may lead me to reinstalling that ATA66 drive as it was previously configured, and see if my results can be duplicated. Too bad these drives aren’t hot swappable! At the very least, I should’ve taken the time to run my PS tests a few times with each drive configuration to get an average performance value.

I didn’t jump on the personal computer (PC or Mac) bandwagon until ’95, so maybe that explains my confusion. 😉 Ha!

Daryl
DM
dave_milbut
Jul 22, 2004
good deal brent.
GA
Gabriel_Ayala
Jul 22, 2004
heh! it’s easy to get hooked on this place. i started out a couple years ago, peeking at a thread here and a thread there. soon i started answering the stuff i knew about pc’s and os’s, then basic to intermediate PS questions. now 2 1/2 years later, i STILL learn something here almost every day.

Yeah, I have a really basic question about…

the personal computer (PC or Mac)

Why is it that the Windows OS version of any personal computer is referred to as the PC and a Macintosh personal computer is known as Mac?
BB
brent_bertram
Jul 22, 2004
Well,
IBM ‘s first 8088 machine, ( 5 slot motherboard, cassette port , no hard drive support in bios ) was the " IBM PC" . From that time on, all the clones ( Compaq, at first, then others & as well as home built machines ) were called "PC compatibles " . Even some , like the Tandy 1000 and 2000, which were somewhat short of compatible. At any rate, I suspect that’s why the Wintel platform computers are universally called "PC’s".

🙂

Brent
GA
Gabriel_Ayala
Jul 22, 2004
Tandy 1000

ROTFLMAO

Wintel

What the heck? If you mean Windows-Intel, that’s entirely different subject I don’t want to started on but personally, I prefer AMD over Intel any day.
GH
Grass_Hopper
Jul 22, 2004
I suspect Wintel came into being before AMD really had their foot in the market. Now the moniker has remained, even though it’s not entirely accurate.
BB
brent_bertram
Jul 22, 2004
In the bad old days, we’d soup up our XT clones with NEC v-20 cpu chips. For a given clock , they’d run about 1.8 x as fast as the Intel 8088 . Using Norton SI under DOS ( of course ) a 4.77 MHZ 8088 was 1.0, an NEC v-20 cpu had an SI of 1.8 . Oh what we did for speed ! <G>

🙂

Brent
GA
Gabriel_Ayala
Jul 22, 2004
I didn’t even get my own computer until they introduced the 200 Mhz with MMX. MMX… woop-dee-doo.

It was my first and last pentium and it was an NEC machine as well. After that, I started building my own machines.
DM
dave_milbut
Jul 22, 2004
pc xt 8088. whoo hoo!
DP
Daryl_Pritchard
Jul 23, 2004
Gabriel,

My first PC was a custom-ordered Pentium 133 with a Number 9 Imagine-128 graphics card…a sweet card at the time. I later upgraded to a P200 MMX on an ABIT IT5H mobo and overclocked it 233MHz, and thought that quite an improvment over the P133. Then I did a major jump to a dual 450MHz P2 in a dual-boot Win98/NT4 configuration. That system went to dual 550MHz P3s just prior to my intalling PS CS, and it continues to serve me until I buy my next "dream system"….unless this "cheapo" Compaq that I’ve just bought to use as a media center changes my mind quickly into deciding I don’t need anything more than it offers for my Photoshop tasks. It’s an Athlon 3000 with now 1GB PC2700 DDRAM and a pair of 160GB drives. But, it’s connected to my 43-inch DLP televsion (it has VGA and DVI inputs) rather than my color-calibrated Iiyama monitor where I’d rather be doing PS edits. For not much over $1000, it kills my dual P3 in performance…about a 500% gain now with one test I’ve run using PS CS.

I’d much rather build a high-performance PC than consider buying anything pre-built, and I rather suprised myself in caving in to buying this Compaq. But, I stripped it of the Compaq fluff, added the extra stick of DDRAM, better sound and graphics, and the extra hard drive, and it still strikes me as a good value.

Daryl
PC
Pierre_Courtejoie
Jul 23, 2004
Dary said:
"With the ATA100/ATA66 mix, I ran a test action on a 22MB file that took 55 seconds to complete in PS CS, with the ATA66 drive serving as the PS scratch disk (with 1GB DDRAM, I’m not sure if the scratch disk even got used…I forgot to monitor it). When the new SATA drive arrived, I installed it and removed the ATA66 drive. So, now I had a single ATA100 drive on an unshared IDE channel and a SATA drive similarly on its own. Rerunning the test action, the time improved to 41 seconds! "
So, the new Scratch disk is a modern sata drive now…
the difference in speed comes mostly from the fact that the scratch disk is now on a modern disk, VS a 3 years old one.

I seriously doubt that the bottleneck would have been the max theorethical transfer rate, as it applies only on the data that sits in the cache. (You ATA 66 had either 512k or 2000k of cache, but your sata now has 8000k of cache… so with two 8mb cache disks, you could begin to see differences with the cache’s data, if the cached info is the one needed.)
The main difference between your two disks is the Sustained Transfer Rate… Photoshop Scratchs consists of huge files, way bigger than the 8mb of cache of the modern hard drives.

If you look at Storage review, see the STR of a ATA 66 GENERATION disk and the STR of a MODERN SATA disk… ouch1 the ATA 66 aren’t there anymore… so one of the oldest ATA100 HD had an STR of 34 MB, while modern stas top at 60MB/S…
DP
Daryl_Pritchard
Jul 23, 2004
Hi Pierre,

I’d wondered if the performance gain I saw was more due to the performance of the SATA drive itself, but as the test file my action ran on was only 22MB, I didn’t think it likely the scratch disk was even being hit much, if any. I should’ve monitored scratch disk usage. Actually, I guess I could at least do that right now with the SATA disk and, if it is getting hit more than I realize, then even without reconfiguring back to the ATA100/66 mix I had at first, I could at least safely assume the scratch disk really is a larger factor. Time to go take a peek…

Daryl
DP
Daryl_Pritchard
Jul 23, 2004
Well, I’ve been playing around a while and can at least say that now that I’ve been more willing to work with a 100MB file size on a fast PC, I see what a difference the Adjusted Refresh plugin makes. I rarely work with files of this size, particuarly not on the old system or, when I have, the edits have been relatively minor and quick to do.

But, this test action I’ve been using runs quickly save for where the Palette Knife and Extrude filters are applied, which I think Chris Cox once told me are really more just exercising the CPU than anything. That is, this action probably doesn’t really stress the memory and drive resources too heavily.

For all practical purposes, running the action on my 26MB test image results in roughly the same performance in PSCS, whether using the SATA drive as the PS scratch or simply using the startup drive. Going back to PS7, the results are also similar…everything hovering around 35-38 seconds and peak scrath memory usage never more than 125MB of 682MB available RAM. So, this leaves me thinking that the scratch disk wasn’t being hit at all and that perhaps the ATA66 drive sharing the same channel with the ATA100 was in fact hampering the peak performance that the ATA100 could deliver. However, this may have been more from an overall O/S operational perspective than just Photoshop. If all system transactions were occurring more slowly on an IDE channel with mixed devices, then that would surely impact PS. I think I’m going to have to go back to that mixed configuration to test out that idea, as maybe my one timing of 55 seconds wasn’t reliable.

To beef up the action a bit, I added an initial step that resizes the test image to 100MB. That’s where I really saw the payoff in using the Adjusted Refresh plug-in, as it reduced the action’s run time from 5 minutes to just under 4 minutes, obviously benefitting from the smaller tile size used (peak scratch size dropped from 1.42GB to 1.2GB). Running the action without a secondary scratch disk cost a 14 second penalty. So, now that running the action on a larger image size is definitely forcing use of the scratch disk, I guess I’ll go revisit the ATA100/ATA66 mixed territory.

Daryl
GA
Gabriel_Ayala
Jul 23, 2004
That’s great and all but you guys are still talking about systems that are yesterdays news.

I’d much rather build a high-performance PC than consider buying anything pre-built, and I rather surprised myself in caving in to buying this Compaq.

I am almost caving into getting an HP but I want the AMD64.

< http://www.microsoft.com/windowsxp/64bit/howtobuy/default.ms px>

With Windows 64 bit you are not talking about 100’s MegaBytes or GigaBytes of RAM in the single digits anymore. You can go 16GB RAM and even into the TeraBytes! SATA is still a great bargain since it performs almost as good as SCSI.

Since I first saw the O/P I knew this was going to snowball into a discussion about super computers. 😉
DM
dave_milbut
Jul 23, 2004
win64 is still in beta (except for the itaniums)
DP
Daryl_Pritchard
Jul 23, 2004
Well Gabriel, if anything, while I can still dream of a truly high-end system, I guess my recent foray into purchasing this Compaq with Athlon 3000 has really shown me that the cost differential for a "dream system" may not be too easily justified. Then again, that’s usually the difference between dreams and reality, right? 🙂

Needless to say however, I’m definitely learning a bit about various hard drive configurations here…just nothing RAID. Hmmm…maybe I need to buy another SATA drive and RAID it with the current one? Next thing ya’ know, I will be using this system for Photoshop use.

Anyway, there was one thing I totally overlooked when I ran my first set of tests where PS CS peformance showed that 15 second difference in execution on the ATA66/100 mix vs. ATA/SATA independent channels…this was a new PS CS installation. I don’t know if thumbnail caching is automatically initiated with the first launch of PS CS or not, but I’m betting that it was caching of thumbnails that may have been hitting my system at the time I had those mixed drives. That makes much more sense, because peformance of the drives notwithstanding, I’d still not have expected to see such a difference for a relatively small image file. I’m still seeing lesser performance now that I’m back to the mixed ATA100/66 configuration, but nothing as drastic. I’ll be back with more numbers shortly.

Daryl
GA
Gabriel_Ayala
Jul 23, 2004
win64 is still in beta

Good, that gives everyone more time to save up more money for a real beast!
DM
dave_milbut
Jul 23, 2004
yup. the intel 64’s’ll be out just before win64! 🙂
DP
Daryl_Pritchard
Jul 23, 2004
As long as I’ve waiting to move from my dual P3 rig to something current, maybe now this new Athlon 3000 will tide me over until 128-bit is the new buzzword in processing and operating systems!

Maybe not…seeing as how this Compaq just shut down suddenly while doing little more than browsing this web page.

Speaking of which…Does anyone know if an insufficient power supply could be the cause of a sudden shutdown? This PC’s power supply is a scant 250W unit of unknown quality (Lite-On) and it’s now supplying power to a pair of hard drives, a pair of DVD burners, and a GeForce 5700 graphics card. The latter spec’ed a 250W supply as a minimum recommendation, and I’m inclined to swap out to a higher quality Antec or similar supply in the 300-400W range. But, as I’m still aiming for this to be a media center PC, I’m still trying to keep it as quiet as possible. The graphics card introduced two small fans that are audible but not objectionable. I figure a smart power supply should be reasonably quiet but am just curious if everyone agrees a higher wattage unit would be smart? Given that I’ve been playing around with PS on this sytem today, maybe that has been causing a higher power draw since both drives were more active, and that in turn triggered the sudden shutdown?

Now…returning to my original topic, here are some time measurements (min:sec) for PS CS execution of my test action on a 26MB image file and for a 100MB (where the first step upsizes the 26MB file) file:

— ATA66 Hard Drive: Quantum Fireball 1ct
— ATA100 Hard Drive: Western Digital Caviar 1600BB (2MB buffer — SATA Hard Drive: Hitachi Deskstar (8MB buffer)
— RAM allocation 66%
— Scratch size on 26MB file: 84.4MB peaking to 372MB
— Scratch size on 100MB file: 226MB peaking to 1.2GB

— For ATA100/66 Mix on same IDE channel with ATA66 as PS scratch 26MB file: 0:42 average for 4 runs
100MB file: 5:01 average for 2 runs

— For separate ATA100 and SATA channels, with SATA as PS scratch 26MB file: 0:36 average for 4 runs
100MB file: 3:54 average for 2 runs

— For ATA100 as both system and scratch (whether ATA66 or SATA installed) 26MB file: 0:39 average for 4 runs
100MB file: 4:14 average for 2 runs

It was interesting to see that use of a single, fast drive for PS CS yields faster performance than adding in a second, yet lower performance drive. The performance differences are obviously larger than I realized between the particular ATA66 and ATA100 drives that I own, but I somewhat would have thought that isolating scratch disk usage apart from a busier system drive would still have performance advantages. That doesn’t seem the case however. I think my original observations were more dramatic for the suspected reason that PS CS may have been doing some background image caching I’d overlooked, and the ATA66 drive simply is much more sluggish than either of my two new drives. I don’t think mixing the ATA66 and ATA100 drives was the factor I first thought.

Regards,

Daryl
DM
dave_milbut
Jul 23, 2004
It’s hard for me to gauge becasue i went from a p3 800 w/384 ram and ata66 to a p4 hyperthreaded 2.8g w/a gig and a sata (150). all i know is, "me likes". 🙂
WH
Wade_Halva
Jul 23, 2004
I would say that the power supply in that box needs to be upped. I was running into power issues with my old system – similar to your old system, dual CPU, dual hd, dual cd, etc…New power supply, no problems.

HTH,
Wade
DP
Daryl_Pritchard
Jul 24, 2004
Hi Wade,

Yes, thanks, yours is the 2nd opinion I’ve heard today (the other being a tech rep at Best Buy) that my power supply suspicions were probably correct. I’ll give this system a week or so to see if the shutdown occurs again and prepare a possible power supply order.

Daryl
H
Ho
Jul 24, 2004
PC Power and Cooling makes the best power supplies. Period. (a biased testimonial).
PC
Pierre_Courtejoie
Jul 24, 2004
Ditto, get a bigger power supply!

If there is something that most people are not aware, it is how loud are most PC components. It is such a pleasure to work with a quiet computer. You wonder how you were able to work next to a jet engine!

Take a look at www.silentpcreview.com , Darryl, you’ll see a lot of advice about quality power supplies: they test their efficiency, noise, and heat production. The more heat often equals the shortest life for most PC components…
DM
dave_milbut
Jul 24, 2004
Ditto, get a bigger power supply!

nodding. and add a couple of fans. can’t go wrong with more cooling. i’m convinced that most major system malfunctions are the result of repeated overheating.
DM
dave_milbut
Jul 24, 2004
The more heat often equals the shortest life for most PC components…

next time i’ll read your ENTIRE post before hitting send! 🙂

<nodding> at that one too.
DP
Daryl_Pritchard
Jul 24, 2004
Thanks all for the suggestions. So far, this new PC seems to be running relatively cool, which I attribute to it having an uncrowded interior. I did notice the drives were rather warm to the touch, but that was more while I had 3 in place to test out the prior stuff discussed here. Now there’s a nice bare slot between the drives, improving their "breathing room".

Dave, I too believe in keeping things cool, but it does have its drawbacks…the way my old tower tends to hover (9 fans inside), things just don’t sit on top of the case very sturdily. 😉

Daryl
DM
dave_milbut
Jul 24, 2004
heh!
ND
Nick_Decker
Jul 24, 2004
Daryl, another vote from me on getting a bigger and better power supply. And, I’ll echo Pierre’s recommendation for SilentPCReview.com and their forums. They’re not just about quiet PCs, but also about powerful PCs that are quiet.

Power supply ratings are often misleading, so you need to do your homework there and make sure that your 400W (or whatever) PSU is also efficient. I hung out quite a bit at the SPCR forums before building my latest box, and ended up buying a Seasonic 400W unit (about $90US) that is one of the most efficient PSUs on the planet. Efficiency equals less electrical consumption and less heat produced, both good things.

There are several free utilities available for monitoring your various temps. If you’re interested, I’ll post some links for you.

Nick
PC
Pierre_Courtejoie
Jul 24, 2004
Nick, thanks to us there will be no more flamewars in here, we will make so many converts that it would be tto cold and silent for that 😉
DP
Daryl_Pritchard
Jul 25, 2004
Hi Nick,

I saw a Seasonic, I’m pretty sure, at Fry’s Electronics tonight but wasn’t sure of the brand or quality. I don’t recall the price, but I did wind up going ahead and buying a new supply…an Enermax "Noisetaker" rated at 420W total. I’ve seen some good reviews of Enermax and Antec power supplies, and this one seemed to be priced reasonably well at $77, only $8 more than Newegg pricing. I like the idea of dual 12v rails, although if you have a well-built and regulated supply, that might be more just an extra marketing gimmick. Compared to a few other units in the store that I looked over, this one seemed very solidly built, with shielded cabling and hefty enough to suggest a better than average transformer.

I’ve got SiSoft SANDRA 2004 and use it occasionally to check voltages, temps, etc. I’ve also downloaded a couple other monitoring utilities, HMonitor and SpeedFan, which I’ve yet to try out. I’m not too concerned with routine monitoring but if you know of some other, perhaps better such utilities, I’m game to try them.

Thanks,

Daryl
ND
Nick_Decker
Jul 25, 2004
Pierre, as you know, quieting a computer can become as much of an obsession as PS itself! <g>

Daryl, the Enermax Noisetaker PSUs get excellent marks, from what I’ve read. I haven’t seen a review of the 420W model, but the efficiency of the 475W model is right up there among the best.

You can find links to some monitoring utilities here < http://www.silentpcreview.com/Web_Links+index-req-viewlink-c id-19.html>. The main one that I use is Motherboard Monitor, mainly to keep track of my CPU temp from within Windows. I also use DTemp to occasionally check HDD temps. I wouldn’t generally be concerned with these things, but I’ve recently built a new box (first time!), am still tinkering/fine tuning and want to be sure everything is working as it should.

Nick
DP
Daryl_Pritchard
Jul 25, 2004
Hi HIck,

It’s fun building a PC, especially when followed with the satisfaction of seeing everything perform better than many off-the-shelf systems. When I bought this Compaq, I thought that I was getting a pretty nice little system for the purpose intended and for the price. However, having built a system before, it didn’t take long at all for me to want to "dress it up a bit", and now I’m wondering if I’ve saved any money over what I’d have done if building the full system from the outset and just keeping my price objective in mind.

What I started with was an Athlon XP 3000 on an ASUS fully-integrated mobo with only 3 PCI slots. What I liked was the simple look of the black Compaq case, with digital media, 3 USB2, and 1 firewire port in front. Add in that it also provided a 160GB hard drive, CD drive, 8X DVD+R HP burner, and 512MB DDRAM (albeit PC2700), and the $579 price after rebate seemed pretty good. But, I’ve now added another 512MB DDRAM stick, Audigy 2 sound card, Albatron GeForce FX 5700 graphics card, 160GB SATA hard drive, NEC DVD+R Dual Layer/DVD-R drive (replacing the CD drive), Logitech MX Bluteooth keyboard/mouse combo, and now the new Enermax power supply. This brings the grand total up into the $1200 region for which I ask what value was gained, if any, in buying the Compaq as a starting bade? Not much…lower-end RAM and a mobo that is less fleixible than most. But, just nibbling here and there for the new add-ons just didn’t feel as expensive as spending a larger chunk of cash of front. And, I really didn’t intend to do all this…but I should’ve known I would…I just can’t be satisfied with a low-end system even when the intended use could likely have been met with just the original purchase save for the added sound card. Oh well.

The Enermax power supply does seem to have added a small amount of extra noise, but the overall sound from the multiple fans in the system still doesn’t bother me, even when the room is quiet. They’re inaudible once the stereo or TV is on. The only negative I see with the Enermax comes from what is also good about it…the EMI-shielded cable that looks so clean and neat is also quite stiff and thus a bit harder to position inside a case where there’s not much room between the backs of the optical drives and the power supply. But, all is in place now. With the limited mobo, I can’t take advantage of fan monitoring and control for the power supply, as while the p/s provides a connector for that, the mobo doesn’t have the socket. Also, the p/s has a manual fan speed adjustment but I can’t tell any difference in noise volume as the result of using it.

While I’ve never been much of a gamer, given that the monitor for this PC is my widescreen TV, I may find I enjoy some games every now and then…which I guess explains the final expense last night of buying a wireless joystick. Ha! Wireless everything is sweet. 🙂

Daryl
ND
Nick_Decker
Jul 25, 2004
Hey, who you callin’ "HIck", pal? <G>

Sounds like a good system, Daryl. Have fun!
DM
dave_milbut
Jul 25, 2004
warning for newbies… many of those utils can be dangerous to the health of your system! make sure you know what you’re doing before running them. the monitors are ok, but i’d be leary about anything that’s directly messing with my mobo voltage and fan speeds on a generic level (ie. not directly made for your specific hardware).
DP
Daryl_Pritchard
Jul 25, 2004
Ooops!

Sorry about that typo Nick! I was up until 3:30am this morning and then got up at 7am when my dog was barking through the window at a rabbit. Curiosity about a problem I started having with a USB key found me back at the PC rather than back in bed, and addiction to the forum followed shortly afterwards, leading me to adding my reply in a half-dead state. I should’ve checked my spelling, but didn’t. Or maybe I’ll just blame it on this normal style Logitech keyboard…the only thing I gave up in exchange for Bluetooth was the natural, split layout of my MS keyboard which I really, really like.

For what it’s worth, should anyone be curious, I bought a Sandisk Cruzer Mini USB 2.0 flash key that became unreadable for no reason after 3-4 uses. Trying it in my two other PCs gave the same result…detected but unreadable. I exchanged it at Best Buy for another that gave the same problem fresh out of the box. Searching the SanDisk support site, I found similar problems where the response provided involves editing the registry to delete related Sandisk entries, rebooting, then installing the key again. Well, I’ve found that I can’t delete all of the cited keys, even when booting into safe mode in WinXP….that seems rather odd. I found two sets of persistent keys that I’m thinking may each be tied to a particular hardware ID for the key used. Several people reported similarly unreadable Lexar USB 2.0 JumpDrive keys that leave me to believe the cause of the problem is similar. If there are persistent, yet possibly corrupt registry keys being written for these devices, then that may mean that folks should stay away from them. Sleepy though I am, I’m still trying to figure out a way to delete those keys. They are older than the oldest restore point on my system, so System Restore won’t help. My next step is to see if I can launch Regedit from the recovery console and delete the keys there.

Daryl
ND
Nick_Decker
Jul 25, 2004
Daryl, no prob on the typo. I’ve been called worse.

8~)
DM
dave_milbut
Jul 28, 2004
just an update to some earlier posts in this thread:

Microsoft delays Windows Server update, 64-bit OS

<http://zdnet.com.com/2100-1104_2-5285913.html>
PC
Pierre_Courtejoie
Jul 28, 2004
Nick, liek stiff necked Nick ? 😉

DAve, I won’t reprint what I said when I read that. I’m suer that it is because they whave to redo some code bacause of the Intel "port" or AMD64.

Please, Chris and other Engineers/ Non Uniform Memory Access is a must!!!
H
Ho
Jul 28, 2004
<chant> NUMA, NUMA, NUMA…
ND
Nick_Decker
Jul 28, 2004
Pierre, yes, "Nick, liek stiff necked Nick"

I really don’t know why. I’d gladly have a drink with you or Len. Oh wait, if I have a drink with Len it’s gonna cost me a ton! I’m avoiding him.

8~)
L
LenHewitt
Jul 29, 2004
Nick,

Fortunately there is no way I could drink all the beers you’ve said you owe me over the years at one sitting <g>…but I could try.

MacBook Pro 16” Mockups 🔥

– in 4 materials (clay versions included)

– 12 scenes

– 48 MacBook Pro 16″ mockups

– 6000 x 4500 px

Related Discussion Topics

Nice and short text about related topics in discussion sections