Puget Quiet Gaming @ [H] Consumer

[Jason_Wall]

Our new article from Scott Unzicker delves into the sensitive world of silent systems. Puget puts a lot of effort into making this system whisper quiet, but they don't cover their bases and make sure that the system is still stable for gaming.

This system is advertised as a “quiet” gaming machine. If the cooling solution could keep up with the rigors of intense gameplay, we’d feel a lot better about giving a whole-hearted thumbs up. As it stands, Puget has some work to do in order to solve the heat dissipation problems. The “solution” they provided us with negated the advertised characteristic of the machine as a “quiet” one. The system did not work as intended straight of the box.

Thanks for reading!

 

[flenser]

Look closer at the nvcpl entry... an entry using rundll to run nvcpl.dll is a legitimate entry for nvidia's control panel. nvcpl.exe is the malware. Since there was no mention of the difference between the two and the system has an nvidia video card, should we assume it's a false alarm and the review will be corrected?

Let me guess... The reviewer has an ATI card in his home system? Any real nvidia !!!!!! would have specified whether nvcpl.dll or nvcpl.exe was listed.

edit - "fan b o i" is a banned word? sheesh Anyhow...

Other than that, I'm a bit suprised they were recommended, even with reservations. They completely failed to deliver a quiet gaming system due to the totally inappropriate video card choice for that cooling setup, and their tech support guy was competent but required persistence to contact. Sounds like an infield pop fly for an easy out, not the base hit you gave them. I couldn't possibly recommend anyone buy that system from them unless they were able to determine for themselves which parts would work together and who are willing and able to do some tech support troubleshooting themselves.

 

[Dark Prodigy]

Excellent read as usual. Though I will say that for the system specs, 2k seemed rather steep, even in comparison to other boutique type builders, not limited to but including VM.

 

[Hotdog453]

I love reading these articles, simply because it gives an awesome insight into what it'd be like to actually buy one of these systems... keep up the awesome work!

That being said, I'll agree with this:

Other than that, I'm a bit suprised they were recommended, even with reservations. They completely failed to deliver a quiet gaming system due to the totally inappropriate video card choice for that cooling setup, and their tech support guy was competent but required persistence to contact. Sounds like an infield pop fly for an easy out, not the base hit you gave them

If I was to spend ~2k on a machine to be quiet, and they told me to turn up the fans to be able to game with it... I'd be slightly pissed. I have to put my mind in a whole different gear, you realize: My current gaming system is LOUD. But my dad, for example, would not want a loud machine. He might actually consider a system like this. Then to totally negate it by turning up the fans? Come now, that's just an easy out. That just reeks of "we never tested any games or 3dmark or rthdribl (HDR Benchy) in shop, and we never knew it over heated, tee-hee ^_^!" kind of issue to me.

Again, though... great article. Compared to every other "system review" out on the web, you guys make them look like babbling retards with credit cards.

 

[GAZZA]

I would have thought that most of those scores would have been reduced if the said fans were running at full speed if that's what it took to help solve the heat issue.

Here's a thought!
What would have happened had a 7600Gt with a 3rd party video card cooler(such as Artic Cooler) along with the fans running at low speed (quiet) and then compare it to the video card with the big heat pipe with fans running at full speed!
I'll put my money on the Artic Cooler!

I would also bet that most if not all OEM's don't even think about the noise aspect of a pc, I wonder if Puget did any noise testing for the said quiet gaming system to see if the hand fits the glove so to speak????

 

[pugetsys]

Hey guys! Thanks for another review. I wanted to point out a few corrections:

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

1) nvCpl is from nVidia's control panel, no spyware here!

2) Your might want to adjust that way you display pricing for these reviews. You paid $1581.81 for the system, and a whopping $170 for express shipping. Shipping is definitely a real cost of the system, but if another customer is closer, or chooses a lighter or smaller system, the shipping costs would obviously be a lot lower. Pricing today is $1689.47 before shipping.

3) It is too bad this review system is from way back in August, because about two weeks after the review system was purchased, we had already implemented measures to allow us to catch overheating issues before machines leave our floor! We now take continuous temperature readings as the systems go through their stress tests. Here is a sample of the temperature graph we also include now in our packets:

Now, keep in mind, Daniel did not suggest the fan at full speed as a solution. Obviously, that would not be a quiet computer! The purpose of the fan at full speed was to determine whether it fixed the problem -- it was to make sure that an overheating video card was in fact the issue. The next step would have been to ramp down the speed and get your system quiet again, but with enough airflow to keep the video card cool.

I'm confident that the problems you had are not from a design flaw. The risks of overheating are definitely present in this design, but managed correctly we've been able to ship hundreds of units nearly identical to this without problems. In fact, this configuration (or the Conroe equivalent) is one of our most popular configurations.

4) We have had 2-3 reports of voicemails not getting through, and they all happened while we were having some phone system issues early last month. What actually happened is that we did get the voicemail, but we heard only static. The reports made it to my desk, and wouldn't you know it, the customer I called to follow up with ended up being HardOCP! I called **********, and asked if he could provide any additional details about the calls we didn't return. I was going to ask him whether he was calling from a cell phone or VoIP system to try to narrow down the cause, but he actually told me that each voicemail was returned promptly, and that he never had troubles. I was confused, but assumed I must have had incorrect information. Unfortunately, it looks like my information was correct after all. As far as we can tell, we only had problems for a 1-2 week period early last month, and only 2-3 customers reported an issue....but after this review I'll be looking into those issues again.

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Thanks again for the review! While in the past you've seen Puget react to reviews by improving itself, it seems this time around we beat you to the punch. Temperature monitoring has already been part of our process for a few months, with great success! I'll be looking into the voicemail issues again, but since the problems were isolated to a few weeks in September, I think we were more likely dealing with a service issue from our phone provider (which I'll certainly be unhappily reporting to them).

Jon Bach - President
Puget Custom Computers
------------------------------------
http://www.pugetsystems.com
[email protected]
(425) 458-0273 x1003
------------------------------------

 

[poley]

I'll just add my voice to the 2 already that NvCpl is a valid entry for systems with nvidia cards in them, although you can't tell from the picture provided in the article, if you had expanded the command column in msconfig you would have been able to see that the full entry is NvCpl.dll and not .exe, which makes all the difference.

I think the reviewer was more than a little harsh on their opinion of the P-180 case, I for one like a case with clean lines and that doesn't stand out with markings all over it, not all of us like big flashy neon lights (I'm exaggerating I know) and grills emanating from our cases...maybe I'm just boring and hence actually like the case...

The issue with the door for the case is also I feel overemphasized as I have had a case with a door for years and never had an issue with it through swapping out dvds/cds. Whether the above actually effected the overall score in the build quality section is questionable considering the problems with over-heating but still a more subjective opinion of the case would have been more fair.

anyways, the above all relate to one small area of the review, everything else was great

 

[Jason_Wall]

After double-checking the NvCpl issue, I can confirm that it is definitely the innocuous .dll file common to essentially all NVIDIA-equipped machines. I was initially assured that it was the .exe file, which is a well-known spybot virus.

Our sincere apologies to the readers, and to Puget for our mistake. The article has been corrected.

 

[Jason_Wall]

Hey guys! Thanks for another review. I wanted to point out a few corrections:

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
2) Your might want to adjust that way you display pricing for these reviews. You paid $1581.81 for the system, and a whopping $170 for express shipping. Shipping is definitely a real cost of the system, but if another customer is closer, or chooses a lighter or smaller system, the shipping costs would obviously be a lot lower. Pricing today is $1689.47 before shipping.

3) It is too bad this review system is from way back in August, because about two weeks after the review system was purchased, we had already implemented measures to allow us to catch overheating issues before machines leave our floor! We now take continuous temperature readings as the systems go through their stress tests. Here is a sample of the temperature graph we also include now in our packets:

Now, keep in mind, Daniel did not suggest the fan at full speed as a solution. Obviously, that would not be a quiet computer! The purpose of the fan at full speed was to determine whether it fixed the problem -- it was to make sure that an overheating video card was in fact the issue. The next step would have been to ramp down the speed and get your system quiet again, but with enough airflow to keep the video card cool.

I'm confident that the problems you had are not from a design flaw. The risks of overheating are definitely present in this design, but managed correctly we've been able to ship hundreds of units nearly identical to this without problems. In fact, this configuration (or the Conroe equivalent) is one of our most popular configurations.

4) We have had 2-3 reports of voicemails not getting through, and they all happened while we were having some phone system issues early last month. What actually happened is that we did get the voicemail, but we heard only static. The reports made it to my desk, and wouldn't you know it, the customer I called to follow up with ended up being HardOCP! I called **********, and asked if he could provide any additional details about the calls we didn't return. I was going to ask him whether he was calling from a cell phone or VoIP system to try to narrow down the cause, but he actually told me that each voicemail was returned promptly, and that he never had troubles. I was confused, but assumed I must have had incorrect information. Unfortunately, it looks like my information was correct after all. As far as we can tell, we only had problems for a 1-2 week period early last month, and only 2-3 customers reported an issue....but after this review I'll be looking into those issues again.

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Thanks again for the review! While in the past you've seen Puget react to reviews by improving itself, it seems this time around we beat you to the punch. Temperature monitoring has already been part of our process for a few months, with great success! I'll be looking into the voicemail issues again, but since the problems were isolated to a few weeks in September, I think we were more likely dealing with a service issue from our phone provider (which I'll certainly be unhappily reporting to them).

Jon Bach - President
Puget Custom Computers
------------------------------------
http://www.pugetsystems.com
[email protected]
(425) 458-0273 x1003
------------------------------------

Thanks for dropping by, Jon, and thanks for your notes.

Indeed, shipping is a key figure in the overall price, which is why we include it in our configuration table. We almost exclusively choose the cheapest shipping option to save everyone money, so the $170 was the option selected by default. One thing we did fail to mention is that we received a discount on our original purchase that almost negated our shipping cost, but also put the system back at its original price. I assume it was for buying a Certified system. The "Current Price" is our best estimate of what the machine costs at the date stated - which sometimes will include shipping fees if they are available in the respective configurator. It's not always accurate to the penny, and without actually purchasing the system again, we can't know for sure exactly how much the machine will cost. The price of the machine has definitely gone up, though, which was our point.

The biggest problem that we run into with these LOOOONG evaluation periods is that the integrator is sure to change something in the meantime. We are thrilled to hear that Puget has looked at the cooling issue and taken steps to address it.

Because our system was not stable at any other fan level than "high," we were unable to move forward with any other option intrinsic to our current configuration. We were essentially stuck with what we had.

We've had some bad luck with v-mail systems. We tend to be calling exactly when the company is having issues. We're glad to hear that they have since been resolved. Because of the way we have to buy systems, the person you talked to was not the one evaluating the system. We have since corrected that for future articles, but he gave you some inaccurate information, so we apologize if it affected your internal investigation of the matter.

Thanks again for coming on! It means a lot for representatives of the company (much less the president) to come on the forums and talk about the experience.

 

[Jason_Wall]

I think the reviewer was more than a little harsh on their opinion of the P-180 case, I for one like a case with clean lines and that doesn't stand out with markings all over it, not all of us like big flashy neon lights (I'm exaggerating I know) and grills emanating from our cases...maybe I'm just boring and hence actually like the case...

The issue with the door for the case is also I feel overemphasized as I have had a case with a door for years and never had an issue with it through swapping out dvds/cds. Whether the above actually effected the overall score in the build quality section is questionable considering the problems with over-heating but still a more subjective opinion of the case would have been more fair.

The case issue was kind of a wash when it came to the score. The excellent wiring pushed our impressions of the build up, along with the clean OS. What brought it down is the overheating problems we had. We realized that our MINOR displeasure with the aesthetics of the chassis is likely a subjective, consumer-specific concern, so Puget was not penalized. Our message was that we typically like to see a company make an integration their own by adding customizations. We realize that some may like just a nice, flat, black system - which is why they weren't hit for it.

 

[Jason_Wall]

Here's a thought!
What would have happened had a 7600Gt with a 3rd party video card cooler(such as Artic Cooler) along with the fans running at low speed (quiet) and then compare it to the video card with the big heat pipe with fans running at full speed!
I'll put my money on the Artic Cooler!

I would also bet that most if not all OEM's don't even think about the noise aspect of a pc, I wonder if Puget did any noise testing for the said quiet gaming system to see if the hand fits the glove so to speak????

Alternatively, a solution like this jobby is another possible solution that would require little time to install and they wouldn't have to worry about having to go through the installation of a dedicated cooler. Cooling on the video card is a relatively easy fix, in our opinion. And it appears that Puget has realized this and has moved to address it.

To be clear, when we received the system, it was absolutely whisper quiet. They certainly succeeded in their original build, and it even completed our productivity testing at low fan speeds. The only areas we ran into trouble are the high-stress applications. Now, if a user is willing to pop into their system guts, turn up their fans for gaming, then turn them back down again afterwards, then they're going to be quite happy with this system. We criticized Puget in our article because we don't think a user should have to do this.

 

[pugetsys]

Indeed, shipping is a key figure in the overall price, which is why we include it in our configuration table. We almost exclusively choose the cheapest shipping option to save everyone money, so the $170 was the option selected by default. One thing we did fail to mention is that we received a discount on our original purchase that almost negated our shipping cost, but also put the system back at its original price. I assume it was for buying a Certified system. The "Current Price" is our best estimate of what the machine costs at the date stated - which sometimes will include shipping fees if they are available in the respective configurator. It's not always accurate to the penny, and without actually purchasing the system again, we can't know for sure exactly how much the machine will cost. The price of the machine has definitely gone up, though, which was our point.

Thanks Jason. That's right, all of our certified systems come with a 10% discount. My only problem is that $170 shipping is not the cheapest option. Here is a breakdown on the pricing of shipping for that system to TX. We charge shipping at cost (no markup or handling fees):

UPS Ground (4 days) $66.03
Fedex Express Saver (1-3 days) $135.14
UPS 3-Day Select $157.15
Fedex 2 Day $174.73
Fedex Standard Overnight 03:00 pm $217.53
UPS 2nd Day Air $242.47
UPS Next Day Air Saver $277.70
UPS Next Day Air $290.19

So, choosing the cheapest option, the price would have been more than $100 cheaper overall. Anyhow, not a huge deal, but I just wanted to make sure your readers knew!

Jon Bach - President
Puget Custom Computers
------------------------------------
http://www.pugetsystems.com
[email protected]
(425) 458-0273 x1003
------------------------------------

 

[oDii]

Great Job as usual [H].

Out of curiosity, where did Puget stick the COA?

I'm also really surprised about the Gigabyte 7600GT - I've built a number (~5) of machines for people with them in and, even when receiving almost no airflow, they've been fine (all stress-tested overnight so I'm pretty confident in that statement) with temperatures well below 80°C primarily. Oh well, goes to show the variability of not only the GPUs but the environment the system is placed in.

 

[theelectic]

I'm also really surprised about the Gigabyte 7600GT - I've built a number (~5) of machines for people with them in and, even when receiving almost no airflow, they've been fine

Same here. I put one in a Dell 8400 that only has one case fan that doubles as a heatsink fan, pulling air from inside the case over the heatsink and outside of the case. Stress tested it with 3DMark06 for 12 hours with no issues, but I know the person using the card is a light gamer and wouldn't play anything 5 hours straight

 

[Geeky1]

At the risk of stating the obvious, the most likely reason the video card was overheating in the configuration the system shipped in is because it was not seeing enough airflow across its heatsinks. Ok... the question then is "why?"

Well, I believe there are a few reasons:
1. The case is missing a fan. P180s ship with all 4 of their fan mounts occupied. Puget must've taken the top fan out of this case (that's the logical conclusion at least, since the P180 has it when it leaves the factory, and it's not pictured in the review), leaving a 120mm diameter hole inches away from the case's sole exhaust fan. I'd ask Puget where they put my fan and why it wasn't in the system...

2. There are holes all over the back of the case-cut in the slot covers and drilled in the space to the side of them.

The "lower" heatsink on the Gigabyte card (the part of the GPU heatsink assembly that takes up the 2nd card slot) opens to the outside of the case for a reason. It's expecting airflow across it generated by either a pressure differential between the case and the room, or by a fan very close to it, such as below the card or on the side panel directly next to it.

Obviously, the P180's fan is not in such a position (though if they'd put the card in the top x16 slot, assuming it fit next to the CPU heatsink, they probably could've avoided this whole problem entirely, though it doesn't look like it would fit) and It's very difficult to build any kind of pressure differential when you have an almost 5" diameter hole in the case and case fans that take in basically exactly as much air as they exhaust.

At their lowest speed, the Antec Tricool fans only push 39cfm (according to Antec). That doesn't generate a particularly quickly moving airstream when it's coming from a 120mm fan. So the front fan wasn't doing much to cool the card, and the rear fan was mostly pulling cool air in from the top of the case and then shooting it right out the back again.

So. Assuming my above contentions are correct, what could Puget have done to make this work properly?

• Left the top fan in the case as an additional exhaust fan, running at low speed
• Replaced the blank plates for the 4 unused card slots with normal solid ones

Those two steps alone may very well have provided enough airflow across the card's heatsink to keep it from overheating. If not, they could have continued with:

• Running one or both of the exhaust fans at medium speed instead of low; that would have brought airflow up to 56cfm per fan for a minimal noise penalty (yes, the fans are 3dBa louder @ medium speed than @ low, according to Antec, but if the Antec fans are accurately rated, even at 28dB they should be nearly inaudible, and certainly not intrusive in all but the quietest settings) or replacing the Antec fans with something that has a halfway decent noise-to-dB ratio, such as these during the initial system build

And, if all the above failed, they could go with blocking the airflow through the large vent to the side of the card slots as a last resort, or even selected a different case that's more able to generate a pressure differential than the P180 is (possibly something like the original Sonata... I haven't played around with a P180 in person though, so that's just a guess), though I doubt it would come to this.

 

[Bbq]

I have several comments about this review.

1. The main page shows the silver P180. The review, the black p180.
2. Inside the Second box, it shows the cardboard cover, and the two foam blocks. Those are default on the P180. Straight from the factory it comes like that.
3. the Silicone gel, is hot glue. In the lower compartment of the p180, those connections come loose rather easily.
4. The idea of the P180 seperating the main hard drives from the rest of the components, is one of the best and worst ideas at the same time. The point of the lower chamber, is to be sealed off from the upper, leaving the hard drives up to the psu and lower chamber fan to be cooled. This in turn, doesn't vent all the hot air from the cpu directly into the psu, leading to the fans spinning up faster in the power supply. The hard drives, same deal. It is designed to be isolated from the hot components.

To Mr. Bach:
Hi, Great system. I have several suggestions for these systems. They are as follows:
1. The fan at the Rear of the upper hard drive compartment. This will bring the fan closer to the GPU, and thus, helping it cool off a bit. I believe the new P180's have this ability, but I am not 100% on this, as I have the older p180.
2. The sealing off of specific rear holes. In order to guide the cooling air from the front upper chamber toward the GPU cooling, instead of straight out the back.
3. Perhaps some form of mounting a low-rpm fan onto the video card? I know that even a small quiet fan spinning at slow speeds will provide significantly improved cooling.

Nice system, and review.

 

[Jason_Wall]

I have several comments about this review.

1. The main page shows the silver P180. The review, the black p180.
2. Inside the Second box, it shows the cardboard cover, and the two foam blocks. Those are default on the P180. Straight from the factory it comes like that.
3. the Silicone gel, is hot glue. In the lower compartment of the p180, those connections come loose rather easily.

We chose the silver chassis for presentation aesthetics reasons. It's easier to see the detail in a small picture with the silver chassis rather than the black. It's the exact same chassis, though, so not the end of the world.

A common misconception is the use of hot glue on connections. Although some integrators use it, those with a decent assembly budget will use silicone-based caulk or gel to secure the connections. As was discussed in our Falcon FragBox article thread recently, glue is a suboptimal solution because it tends to not be totally inert.

 

[oDii]

What could Puget have done to make this work properly?

I don't really believe it's Puget's fault. Gigabyte sell these cards as components of a system, and to me "component" implies "self contained unit" - it doesn't require anything in addition to it to work properly (aside from the obvious motherboard, CPU, RAM etc...). A CPU for instance is sold with a heatsink and fan - everything that is required for that component to function correctly is provided; given the other pieces of the jigsaw are working properly. Simply saying "Oh Puget is completely in the wrong because they didn't turn their fans up" I believe would be incorrect - it does perhaps indicate a level of testing that isn't sufficient given the price paid, but it surely isn't their fault for providing a supposedly quality part that isn't working at it should.

 

[LoneWolf]

At the risk of stating the obvious, the most likely reason the video card was overheating in the configuration the system shipped in is because it was not seeing enough airflow across its heatsinks. Ok... the question then is "why?"

Well, I believe there are a few reasons:
1. The case is missing a fan. P180s ship with all 4 of their fan mounts occupied. Puget must've taken the top fan out of this case (that's the logical conclusion at least, since the P180 has it when it leaves the factory, and it's not pictured in the review), leaving a 120mm diameter hole inches away from the case's sole exhaust fan. I'd ask Puget where they put my fan and why it wasn't in the system...

Actually, P180's have five 120mm fan mounts, but only come with three fans. I just built a system with this exact case. The top, rear, and bottom central fan mounts are occupied. There are two optional front mounts; neither have fans.

Puget's decision to reposition one of the fans (moving the top one to the front) probably helped in quieting the case, since the top fan-mount is likely to convey the most noise to the end user. I have four fans in the build I made (120x38mm panaflo L1A's in top & rear, two Antec Tricools, one in the lower central and one in front) and this build is by no means quiet, but then again, build requirements were to make a system that could dissipate heat in a tropical environment. Since hot air rises anyway, there will still probably be a bit of cooling from air escaping the upper inlet with no fan mounted.

 

[jfreund]

I've built a system on the M2N32-SLI Deluxe and I was able to make it very quiet with parts that generate more heat than those in this Puget system. Granted I used a different case (Coolermaster Mystique), but what got it done for me was not just using lots of fans, but activating the Q-Fan feature in the Asus BIOS for automatic control of fan speeds. At idle, the system I built is virtually indistinguishable from ambient noise. Under load the fan speeds increase, but it never gets to the point where I would say it's loud (and I'm pretty picky about noise).

So, it seems to me that if Puget were to use Q-Fan when configuring the systems at the factory, it could provide a low noise level at idle and yet have adequate cooling under load (when fan noise is less likely to be bothersome anyway).

 

[Fleat]

This is one of the few reviews that I have read and would truely say would have me irate. Noise is an enormous irriation for me which is one of the reasons I love building my own machines. If I purchased a computer that was "supposed" to be capable of doing anything I need WHILE being quiet... that is what I would expect. The moment it started overheating I would have offered to let them fix the problem at there cost (shipping, etc.) in a timely manner or return the machine.

Good review HardOCP, but I think overall if you are purchasing a -silent- machine it should be just that. I would say the value should plummet if it can't function as promised.

 

[lithium726]

I don't really believe it's Puget's fault. Gigabyte sell these cards as components of a system, and to me "component" implies "self contained unit" - it doesn't require anything in addition to it to work properly (aside from the obvious motherboard, CPU, RAM etc...). A CPU for instance is sold with a heatsink and fan - everything that is required for that component to function correctly is provided; given the other pieces of the jigsaw are working properly. Simply saying "Oh Puget is completely in the wrong because they didn't turn their fans up" I believe would be incorrect - it does perhaps indicate a level of testing that isn't sufficient given the price paid, but it surely isn't their fault for providing a supposedly quality part that isn't working at it should.

No. It is Pudget's jop to properly engineer a system that they are selling as a quiet gaming system, which obviously was NOT very quiet with the fans turned up to keep it running. I'm sure the gigabyte card would have run just fine in a properly cooled case.

If you put a prescott in a case with no cooling sans the CPU fan are you gonna blame Intel when it blows up and takes your house with it? Didn't think so. And if you answered yes, I laugh at you.

 

[Geeky1]

oDii... I'm not even sure where to begin addressing that... Nor do I really feel like trying.

Actually, P180's have five 120mm fan mounts, but only come with three fans. I just built a system with this exact case. The top, rear, and bottom central fan mounts are occupied. There are two optional front mounts; neither have fans

LoneWolf... where are the two front fan mounts? I've not been able to play with a P180 yet... (no reason to upgrade from my Stackers...) I know there's a "middle" mount in the front of the case for a 120mm fan-the one Puget moved the top fan to-but where's the other front mount? there's not one in front of the lower hard drive cage is there? I've looked at this case in Fry's before and didn't see one there... I know about the top, the rear, the one between the hdd cage and the power supply, and the middle front mount, but where's the fifth?

As far as hot air rising goes... yes, you're absolutely right. However, convection currents generated by a temperature differential of a few degrees C are not going to be very powerful, especially with such a small amount of air. Having an exhaust fan right next to the now unoccupied fan mount in the top of the case guarantees that the exhaust fan is pulling some air into the case through the top fan mount and pushing it right back out again. If you didn't have that rear exhaust fan, you'd get warm air moving (albeit very slowly) out of both the rear and top fan mounts, but with that fan there and no top fan, it just screws up the airflow through the whole case. Normally this wouldn't matter, but because the video card is entirely passively cooled, it just throws a monkey wrench into the works, as it were.

 

[LoneWolf]

LoneWolf... where are the two front fan mounts? I've not been able to play with a P180 yet... (no reason to upgrade from my Stackers...) I know there's a "middle" mount in the front of the case for a 120mm fan-the one Puget moved the top fan to-but where's the other front mount? there's not one in front of the lower hard drive cage is there? I've looked at this case in Fry's before and didn't see one there... I know about the top, the rear, the one between the hdd cage and the power supply, and the middle front mount, but where's the fifth?

http://www.antec.com/uk/productDetails.php?ProdID=81802

You can mount two fans in the front-middle area, creating a push-pull effect for additional cooling. Note that Antec advises not to do it if you wish for a quiet PC.

 

[oDii]

No. It is Pudget's jop to properly engineer a system that they are selling as a quiet gaming system, which obviously was NOT very quiet with the fans turned up to keep it running. I'm sure the gigabyte card would have run just fine in a properly cooled case.

Puget delivered a quiet gaming system that wasn't tested correctly with a faulty graphics card, which was replaced by another with a faulty graphics card. There is evidence in and below my first post that reinforces the fact that these cards were not operating within their normal thermal envelope, and will work fine as a self contained unit in thousands of other configurations around the world in different thermal conditions - many of which are far less hardware friendly as evidenced by theelectic's post below my first.

I'm not sure why you're arguing with me - obviously not because of my user rank - but to some extent I am agreeing with you; it is Puget's fault for not testing the card/s properly before shipping them to a customer, but it is ultimately Gigabytes fault for a faulty card. We've already established that these cards are capable of working in any environment reasonably expected of them (far worse than the situation in Puget's case), so clearly the cards tested were faulty. Why are you blaming Puget for that?

If you put a prescott in a case with no cooling sans the CPU fan are you gonna blame Intel when it blows up and takes your house with it? Didn't think so. And if you answered yes, I laugh at you.

Do I need to address this? Guess I'll entertain you... in my initial post I deliberately said:

A CPU for instance is sold with a heatsink and fan - everything that is required for that component to function correctly is provided

because I suspected someone would be clever. If it blew up with the CPU fan on, I'd be expecting Intel to answer some serious questions about how their cooling system wasn't fit for its purpose... just like I'd be expecting Gigabyte to replace these cards for Puget because their cards didn't work as promised.

 

[Geeky1]

We've already established that these cards are capable of working in any environment reasonably expected of them (far worse than the situation in Puget's case), so clearly the cards tested were faulty. Why are you blaming Puget for that?

These cards would probably be better off in a Dell with a single exhaust/cpu fan than they would in the P180 in the configuration Puget shipped this system in. I thought I explained fairly clearly why this is in my first post.

 

[lithium726]

Do I need to address this? Guess I'll entertain you... in my initial post I deliberately said:



because I suspected someone would be clever. If it blew up with the CPU fan on, I'd be expecting Intel to answer some serious questions about how their cooling system wasn't fit for its purpose... just like I'd be expecting Gigabyte to replace these cards for Puget because their cards didn't work as promised.

Then I laugh at you. Simply having a CPU fan over a heatsink isnt going do a damn bit of good in a case where there is a shitload of heat building up, you need to have cool air circulating through those fins.

I'm not sure why you're arguing with me - obviously not because of my user rank - but to some extent I am agreeing with you; it is Puget's fault for not testing the card/s properly before shipping them to a customer, but it is ultimately Gigabytes fault for a faulty card. We've already established that these cards are capable of working in any environment reasonably expected of them (far worse than the situation in Puget's case), so clearly the cards tested were faulty. Why are you blaming Puget for that?

If the card was faulty, thats cool. I didnt read the whole thing, really. I was under the impression (from responses and temp readings) that Pudget's cooling system was NOT up to snuff. I was responding to your theory that it is not Pudget's responsiblility to deliver a properly engineered computer with a working cooling system. Without airflow, just about everything in that thing is going to die. Hard drives die when they get too hot, CPU's die or downclock (these days at least), video cards overheat and die, memory deveolps errors, etc. These components are MADE to be put in a properly cooled case and when they are not used in their intended enviroment, they fail. Its not the component manufacturer's fault. Everything in a computer (or any digital system for that matter) will fail when run under hot circumstances, heat is the ultimate enemy of all digital circutry... why do you think companies have air conditioners reserved *only* for server rooms? They should just work right out of hte box in a crowded room with all the other rackmounts, right? Wrizzong.

And the prescott stab was supposed to be a joke. A little cleverness never hurt anyone, you know. Everyone is too damn serious on these boards

as an aside, I dont see how user rank would keep me from arguing with you... I argue with everyone.. and usually win

 

[oDii]

Then I laugh at you. Simply having a CPU fan over a heatsink isnt going do a damn bit of good in a case where there is a shitload of heat building up, you need to have cool air circulating through those fins.

I'll agree that isn't the best case scenario, but the assumption that everything needs fans constantly filling the case with cold air is wrong. Yes, you're not going to get the best temperatures, but expecting it to run properly isn't an unrealistic expectation. That should be the case with graphics cards - in the ATX formfactor they're lucky to get some "prewarmed" air blown over them from the harddrives' intake, and should be designed as such.

I was responding to your theory that it is not Pudget's responsiblility to deliver a properly engineered computer with a working cooling system.

I'm not denying it is - one hopes that they built numerous test machines before releasing these to the consumer - but when you've tested with all fully functional, working cards then suddenly the system is then expected to cool outside the parameters it was designed for (ie. a faulty card), it's the fault of whatever changed those parameters - in this case, Gigabyte.

Its not the component manufacturer's fault. Everything in a computer (or any digital system for that matter) will fail when run under hot circumstances, heat is the ultimate enemy of all digital circutry...

You will decrease the expected lifetime of the parts, yes. But will you kill it immediately? No way. The MTBF of a part, usually set a very unlikely temperature, gives a very optimistic expected lifetime of the parts; statistically, increasing the temperature might kill more hard drives - but is that increase going to massive? No, but that's the tradeoff silencers will make.

you think companies have air conditioners reserved *only* for server rooms? They should just work right out of hte box in a crowded room with all the other rackmounts, right? Wrizzong.

This is a discussion about consumer oriented goods, hence "[H] Consumer". I'd expect server components to have different lifetimes; designed under the expectation of different environmental variables. But you know what? Even when stuck in a non-conditioned room I'd still expect a server to last - perhaps not as long or I at least wouldn't trust it as much as one I'd have in a server conditioned room. But ultimately it's about sacrifice and getting what's most important to you - SPCR members for example will endure slightly hotter components and consequentially perhaps slightly lower lifetimes of computers in the search for lower noise levels, while a gamer who wears headphones won't really care if his PC is noisy if it has higher levels of reliability, stability and the headroom for the possibility of overclocks.

These cards would probably be better off in a Dell with a single exhaust/cpu fan than they would in the P180 in the configuration Puget shipped this system in.

That's somewhat irrelevant given my argument above (card producing more heat than expected/designed for).

Guys, I can see where you're coming from, but in the consumer environment, manufacturers can't make any presumptions about the thermal conditions that their products will be in - hence while it is lax of Puget not to foresee or correct these temperature problems with their products (as they apparently have done now), it is far more Gigabyte's fault for shipping cards that are clearly have the potential to operate outside their operational thermal envelope (ie. heatsink not attached correctly so shipping could dislodge slightly from the thermal glop, or a screw unscrewed that reduced contant with the heatpipes). Any arguments about Presshots are irrelevant; this is a faulty graphics card with different thermal requirements and Puget cannot be blamed for a single faulty component in what is a system designed for a non-faulty card. Puget made an assumption about the TDP of the card that proved incorrect when the card isn't working as it should. As we know: assumptions are the mothers of all screw ups, but this assumption shouldn't have had to factor in faulty cards as it's Gigabyte's responsibility to make reliable cards, and as proven above when the cards are working properly they work fine in similar circumstances.

I am not saying that this is a golden rule, these are simply my expectations and as such I wouldn't be jumping on Puget's back as I am glad to see [H] hasn't - even in the best conditions things will go wrong; things bloody well always do ...

 

[lithium726]

I'll agree that isn't the best case scenario, but the assumption that everything needs fans constantly filling the case with cold air is wrong. Yes, you're not going to get the best temperatures, but expecting it to run properly isn't an unrealistic expectation. That should be the case with graphics cards - in the ATX formfactor they're lucky to get some "prewarmed" air blown over them from the harddrives' intake, and should be designed as such.

You need to stop taking what I say out of context. I never said you had to have constant cold air, I said there needed to be airflow. Obvoiusly there is not going to be a constant supply of "cold" air, but there needs to be some movement of air. When I gave the prescott example I was giving a completly stagnant case scenerio, which is what you are expecting these devices to run under, which is completly unrealistic.

I'm not denying it is - one hopes that they built numerous test machines before releasing these to the consumer - but when you've tested with all fully functional, working cards then suddenly the system is then expected to cool outside the parameters it was designed for (ie. a faulty card), it's the fault of whatever changed those parameters - in this case, Gigabyte.

If that is what happened, that is just peachy. This is no longer about Pudget, but about your theory of self-contained parts being able to work on their own regardless of conditions and I will treat it as such.

You will decrease the expected lifetime of the parts, yes. But will you kill it immediately? No way. The MTBF of a part, usually set a very unlikely temperature, gives a very optimistic expected lifetime of the parts; statistically, increasing the temperature might kill more hard drives - but is that increase going to massive? No, but that's the tradeoff silencers will make.

Yes way. Back in the day if you took a Heatsink off of a processor it died right then. Obvoiusly there are measures now to keep that from happening, but if these chips did not have any thermal throttling (and ask yourself, why do they need thermal throttling? becuase some system builders put inpept cooling systems in there or the consumer does something retarded like put it in a sealed area..) they would die from getting to hot, period.

This is a discussion about consumer oriented goods, hence "[H] Consumer". I'd expect server components to have different lifetimes; designed under the expectation of different environmental variables. But you know what? Even when stuck in a non-conditioned room I'd still expect a server to last - perhaps not as long or I at least wouldn't trust it as much as one I'd have in a server conditioned room. But ultimately it's about sacrifice and getting what's most important to you - SPCR members for example will endure slightly hotter components and consequentially perhaps slightly lower lifetimes of computers in the search for lower noise levels, while a gamer who wears headphones won't really care if his PC is noisy if it has higher levels of reliability, stability and the headroom for the possibility of overclocks.

erm... yeah, guess what? The same components that go into servers go into consumer machines. They all require the same thing, decent temperatures. If you expect a server farm to be just fine in a non-climate regulated room you would be an IT departments worst nightmare.

My computer is silent. I cannot hear it. I have an overclocked P4, an overclocked 6800 series card, overclocked memory, and a small midtower case. Guess what? All my stuff runs well within spec, because I know how to keep everything cool.

Guys, I can see where you're coming from, but in the consumer environment, manufacturers can't make any presumptions about the thermal conditions that their products will be in - hence while it is lax of Puget not to foresee or correct these temperature problems with their products (as they apparently have done), it is far more Gigabyte's fault for shipping cards that are clearly have the potential to operate outside their operational thermal envelope. Any arguments about Presshots are irrelevant; this is a faulty graphics card with different thermal expectations and Puget cannot be blamed for a single faulty component in what is a nicely designed system otherwise.

Absolutly they can. They need to presume the worst. They need to take into account the lowest common demoninator. Our school systems are doing it, why doesnt everyone? This computer was not even subjected to the worst conditions it could have been subjected to and it overheated and failed [H]consumer's tests... they test for the consumer. They dont test for the enthusiast or the server, they test for the consumer and this system failed the consumer level test, which is unacceptable.

and guess what? The stock coolers on ANY card will get FAR too hot in horrible thermal conditions. With no airflow, just about any card will overheat. same goes for processors, which is why Intel and AMD HAD to implement thermal throttling, to save people from being idiots and destroying their stuff and then complaining to the manufacturer, which is what you seem to be suggesting.

I am not saying that this is a golden rule, these are simply my expectations and as such I wouldn't be jumping on Puget's back as I am glad to see [H] hasn't - even in the best conditions things will go wrong; things bloody well always do ...

Of course shit happens. Thats why I said if the card was a reject, thats just fine. That is not their fault. However, the assumption that a component comes with everything it needs to run selfcontained on the card and not taking into account external variables is incredibly ignorant.

That's somewhat irrelevant given my argument above (card producing more heat than expected/designed for).

No, it's not. Fluid dynamics is a real thing, you know, and thier configuration is far from optimised. Geeky may be a "noobie" here, but he definatly knows his shit.

 

[oDii]

Look mate, while you're incredibly good practice in debating, I'm not going to argue forever with you.

I said there needed to be airflow. Obvoiusly there is not going to be a constant supply of "cold" air, but there needs to be some movement of air. When I gave the prescott example I was giving a completly stagnant case scenerio, which is what you are expecting these devices to run under, which is completly unrealistic.

Not a believer in convection?

Yes way. Back in the day if you took a Heatsink off of a processor it died right then. Obvoiusly there are measures now to keep that from happening, but if these chips did not have any thermal throttling (and ask yourself, why do they need thermal throttling? becuase some system builders put inpept cooling systems in there or the consumer does something retarded like put it in a sealed area..) they would die from getting to hot, period.

Are you sure that example is extreme enough?

erm... yeah, guess what? The same components that go into servers go into consumer machines. They all require the same thing, decent temperatures. If you expect a server farm to be just fine in a non-climate regulated room you would be an IT departments worst nightmare.

Read my post again. I would not expect the same from it.

They dont test for the enthusiast or the server, they test for the consumer and this system failed the consumer level test, which is unacceptable.

No, it's not. Fluid dynamics is a real thing, you know, and thier configuration is far from optimised.

The entire point I'm making is that this card is not a good example of the cooling solution used; in this case you're trying to put out a forest fire with a garden hose - sure, it might work when it starts snowing, but in any other conditions you can't possibly expect it to work. This garden hose is not Puget's best work, as normally they'd be fighting a pan fire.

 

[oDii]

This will be my final post in this topic - sorry man, I don't want to be drawn out into a long, time wasting argument.

It's my belief that everything will work in an enclosed environment. Not well; it'll throttle, and the lifetime of the system will be affected. But it will work presuming everything is working correctly and isn't broken (ie. not this Puget example), the system will function for a reasonable amount of time.

Consider an ATX case. Given convection, most hot air will be carried out of the PC but the power supply fan - and yes, while this is a horrible thermal situation, it will still work. Hence with the bare minimum number of fans (which can be assumed by the manufacturer of the components), it should still operate - again, not reliably; no body in their right mind would operate a system like this, but this was the case right up until a few years ago. Look at any pre-2002 standard case and there's barely an exhaust to be seen - 60mm fan holes were a luxury and the PSU was assumed to do the work. While this may not work in the most extreme examples today, it'd be achievable with lower powered components - like the ones in the Puget today. I'll even go half way with you and say that case fans are important, but not critical. Correctly functioning components shouldn't need them as a matter of fact. Yes, I have fans in all the PCs I build and even in my personal machine - as my clients, myself included, all want the machines to be reliable and are not too worried about noise.

If we're going to be pedantic, then any component I accept will not be a "self contained unit" in certain circumstances - I've been to Thailand and endured the 50°C it can commonly reach on the ground and yes, you could not expect any component to work there without assistance. However, given our more moderate climates, it shouldn't be a fact of life - it should be like I said before, a trade off between priorities.

 

[Geeky1]

oDii: hopefully you're still reading, at least. What I'm trying to explain to you is that it is unreasonable to expect them to function in a system with the configuration of the one Puget shipped because the requirements for the card to function are not being met. Failing to provide the card with sufficient airflow is Puget's fault, not Gigabyte's. No current moderately powerful video card can be cooled by convection alone (unless it's put in a case like one of Zalman's TNN cases), period. Part of the requirements of running one of these cards is providing it with the environmental conditions necessary for its operation. And Puget did not do that, as far as I can tell.

There is of course always the possibility that the heatpipes themselves on the two cards were defective (either through a leak or not being filled with the correct amount of working fluid) and it would be the responsibility of Gigabyte's heatsink supplier to catch this. If they didn't, Gigabyte certainly should've, and if Gigabyte didn't, Puget should have. A defect of this kind wouldn't make it through three tiers of quality control if the qc inspectors were paying any kind of attention. So even in this case I'd say Puget carries some level of responsibility, although less so than they would in the more likely situation (which is not providing the card with operating conditions conducive to its operation).

Anything you put in a computer has certain environmental conditions that need to be met for it to function. Expecting components to be self-contained units is unreasonable and unrealistic.

 

[Jim Hall]

Bump or excellent review and keeping em honest!

jim

 

[oDii]

This is my last post. Honestly.

oDii: hopefully you're still reading, at least. What I'm trying to explain to you is that it is unreasonable to expect them to function in a system with the configuration of the one Puget shipped because the requirements for the card to function are not being met.

I disagree. For joe-blow consumer, this is simply a graphics card without a fan - he's not going to know that it needs special care and attention to function because some gimp down at where ever he bought it from is similarly stupid. Then, when installed, if Joe's game of 3D solitaire doesn't work properly, he's going to take it back and say "I've got a broken graphics card". It would be fundamentally idiotic of Gigabyte to make a card that requires something that cannot be relied upon; they're not going to produce such a card as in doing so it would increase the returns of their product. It might be unrealistic and irresponsible to expect this, but it's exactly what Joe-blow consumer is going to expect and this is exactly how any manufacturer should make their cards - to the expectations of the consumer. It would be worth noting here that they do, as my (and others) previous experiences with the card have been nothing but positive, with the cards functioning perfectly with less airflow than in this situation.

There is of course always the possibility that the heatpipes themselves on the two cards were defective (either through a leak or not being filled with the correct amount of working fluid) and it would be the responsibility of Gigabyte's heatsink supplier to catch this. If they didn't, Gigabyte certainly should've, and if Gigabyte didn't, Puget should have. A defect of this kind wouldn't make it through three tiers of quality control if the qc inspectors were paying any kind of attention. So even in this case I'd say Puget carries some level of responsibility, although less so than they would in the more likely situation (which is not providing the card with operating conditions conducive to its operation).

I have no argument with this - everyone is to blame, but I believe that Gigabyte is more to blame as you expect a working, quality product from them. While there is a similar expectation from Puget, we've already established that they need to (and probably already have) improve their testing procedures but ultimately they're at the mercy of their suppliers and shippers - if either get it wrong, they're blamed for it, hence me not wanting to put more blame on them.

Anything you put in a computer has certain environmental conditions that need to be met for it to function. Expecting components to be self-contained units is unreasonable and unrealistic.

No, expecting components not to work in conditions they should have been designed for is unreasonable and unrealistic. I hate to get legal here, but in New Zealand we have something called the CGA (Consumer Guarantees Act) which states that goods must be fit for their purpose. Anyone could argue that if a card doesn't work in a similar situation they use it in then it isn't fit for its purpose as the manufacturer should not assume anything about the airflow in ones system. When you buy a fanless graphics card you (perhaps not you, but joe-blow consumer) should be able to assume similar cooling abilities to a fanned card - ie. it'll look after itself as the differentiation just isn't there for the consumer. These aren't enthusiast level cards, the 7600GT is right in the mainstream's butterzone - so you can't expect enthusiast levels of cooling or knowledge.