Seasonic shows X-Series, modular, 80 Plus Gold @ Computex
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee, Devonavar
-
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
jessekopelman, dhanson865, and everyone:
PSU positioning and cooling configuration in a case has a very significant impact on the (fan)rpm-to-output load curve. Compared to SPCR test results, depending on which PSU, I'd guesstimate that the hinge point on a good PSU noise/load curve could be pushed right (or up) perhaps 100W or more.
The SPCR PSU test rig was designed at a time when there were no cases that provided an independent cooling zone for the PSU; almost all PC cases simply had the PSU on the top back, as per ATX tower spec, sucking in from above the CPU/GPU area and blowing it through the PSU. The hot air flow through the PSU got even worse and more direct when 120mm fan PSUs came on the scene. The SPCR hotbox tester was a good replication of a thermally realistic PC environment for the PSU.
Things have changed. Since I first started doing PSU reviews some 7 years ago, the number of P180-esque cases with PSU intake vents far from the CPU/GPU heat has grown in a big way. Especially recently. Antec has P183, P193, mini P180, 1200, 901, 300, 200. Silverstone has RV01, RV02, FT01, FT02, TJ10, TJ09, PS02. Coolermaster has Centurion 590, cm690 (3 v.), Gladiator 600, HAF922, HAF932, acts840, Cosmos (2 v.). Even Thermaltake has a handful - v9 (2 v.), Element (3 v.) Level 10, Spedo. Lian-Li: Armorsuit PC-P50R, pcx1000, PC-A71, PC-A10, etc..... Among HTPC cases, the concept seems to be implemented even more widely these days.
The point is this: The DIY aftermarket case makers get it. Whether through simple imitation chasing the P180's market success or a real understanding of the technical benefits, it doesn't really matter: Cases that provide cooler air for the PSU are now pretty easy to find.
In this context, SPCR PSU reviews may no longer be quite as uniquely useful as they once were. It may actually be preferable to test with the intake pulling normal room air, because that's the type of case most silencers can/should/would choose.
It might actually be time to stop testing PSUs in the hot box.
I would certainly not entertain fan mods in one of the very quiet PSUs; I'd simply make sure it never got exposed to high enough temperature for the fan to ramp up... by using a case with a separate air intake vent for the PSU.
PSU positioning and cooling configuration in a case has a very significant impact on the (fan)rpm-to-output load curve. Compared to SPCR test results, depending on which PSU, I'd guesstimate that the hinge point on a good PSU noise/load curve could be pushed right (or up) perhaps 100W or more.
The SPCR PSU test rig was designed at a time when there were no cases that provided an independent cooling zone for the PSU; almost all PC cases simply had the PSU on the top back, as per ATX tower spec, sucking in from above the CPU/GPU area and blowing it through the PSU. The hot air flow through the PSU got even worse and more direct when 120mm fan PSUs came on the scene. The SPCR hotbox tester was a good replication of a thermally realistic PC environment for the PSU.
Things have changed. Since I first started doing PSU reviews some 7 years ago, the number of P180-esque cases with PSU intake vents far from the CPU/GPU heat has grown in a big way. Especially recently. Antec has P183, P193, mini P180, 1200, 901, 300, 200. Silverstone has RV01, RV02, FT01, FT02, TJ10, TJ09, PS02. Coolermaster has Centurion 590, cm690 (3 v.), Gladiator 600, HAF922, HAF932, acts840, Cosmos (2 v.). Even Thermaltake has a handful - v9 (2 v.), Element (3 v.) Level 10, Spedo. Lian-Li: Armorsuit PC-P50R, pcx1000, PC-A71, PC-A10, etc..... Among HTPC cases, the concept seems to be implemented even more widely these days.
The point is this: The DIY aftermarket case makers get it. Whether through simple imitation chasing the P180's market success or a real understanding of the technical benefits, it doesn't really matter: Cases that provide cooler air for the PSU are now pretty easy to find.
In this context, SPCR PSU reviews may no longer be quite as uniquely useful as they once were. It may actually be preferable to test with the intake pulling normal room air, because that's the type of case most silencers can/should/would choose.
It might actually be time to stop testing PSUs in the hot box.
I would certainly not entertain fan mods in one of the very quiet PSUs; I'd simply make sure it never got exposed to high enough temperature for the fan to ramp up... by using a case with a separate air intake vent for the PSU.
I don't know. With the increasing emphasis on low power components, it is becoming more practical to have a system cooled by a single low-speed fan. Since the PSU probably has to have a fan anyway, one ideal solution is to have the PSU fan cool the CPU directly.MikeC wrote:...It may actually be preferable to test with the intake pulling normal room air, because that's the type of case most silencers can/should/would choose...
-
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
Sure, but such a low power application would not make any of the top PSUs ramp up at all, so it doesn't require hot testing. Personally, I still prefer a separate fan blowing around the CPU, even if at very low airflow. Components last longer and there's usually no noise penalty with a second quiet fan.Mr Evil wrote:I don't know. With the increasing emphasis on low power components, it is becoming more practical to have a system cooled by a single low-speed fan. Since the PSU probably has to have a fan anyway, one ideal solution is to have the PSU fan cool the CPU directly.MikeC wrote:...It may actually be preferable to test with the intake pulling normal room air, because that's the type of case most silencers can/should/would choose...
-
- Posts: 1406
- Joined: Tue Feb 13, 2007 7:28 pm
- Location: USA
I think you misunderstand me a bit. It's not so much how slow the swappers new fan is, but how unnecessarily fast the old fan would have been. Now when I say unnecessarily, I mean from the point of view of a typical user who just wants it to work and last maybe 4 or 5 years. The engineer who specs the fan has much higher standards than that, so he is going to spec a fan that is serious overkill (from the user perspective). So, again, I think the only warning that need be given is that one is almost certainly shortening the functional lifespan of their PSU by doing a fan swap. Given that the useful lifespan of a too-noisy PSU is 0, I think that is a good trade off. I'd rather have a PSU that is functional and near-silent for 3 years, than a PSU that is functional and noisy for 6+ years.dhanson865 wrote:I was assuming that a SPCR regular that does a fan swap to a fixed 5v PSU fan is going to fix that fan at a very low RPM.
-
- Posts: 5
- Joined: Wed Sep 09, 2009 7:59 am
- Location: USA
-
- Posts: 92
- Joined: Tue May 29, 2007 9:25 am
- Location: Norway
It's been mentioned before, but manufacturers stay in the higher wattage area, because producing 300W PSUs cost about as much as producing a 400-450W PSU and since very few people will actually go out and buy that 300W or even 400W PSU, there's no money to be made. So they'd rather sell those higher wattage (500-650W) PSUs than be stuck with something that they'd make no money on.need4quiet wrote:650W? What the hell are most people going to do with that? Where's the reasonably priced 300W modular power supply for people who leave their mother's basement, don't sit around all day playing video games and don't need more than 150W ever?
-
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
Yes, this has been discussed extensively throughout this thread.Joe Public wrote:It's been mentioned before, but manufacturers stay in the higher wattage area, because producing 300W PSUs cost about as much as producing a 400-450W PSU and since very few people will actually go out and buy that 300W or even 400W PSU, there's no money to be made. So they'd rather sell those higher wattage (500-650W) PSUs than be stuck with something that they'd make no money on.need4quiet wrote:650W? What the hell are most people going to do with that? Where's the reasonably priced 300W modular power supply for people who leave their mother's basement, don't sit around all day playing video games and don't need more than 150W ever?
Note that the 80 Plus test report of the 550W Seasonic gold shows its 20% load efficiency as 87%. That's at 113W output; it only draws 129W. Compare that to my M12D-850 test results, which showed 82% at 90W, 84% at 150, 87.5% at 200W. Extrapolating downward, I'd guess the 550W model could have 80% efficiency at 50W or lower. That's as good as you're going to get from a "conventional" ATX12V PSU.
-
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
An early pre-production sample of the 750W model but I am under NDA to stay mum. I've been told retail samples will be here by the end of the month. It is very impressive, clever technology and robust build.CA_Steve wrote:Mike - has Seasonic provided a sample, yet...or is one expected sometime soon?
Hellllllllllll yeahhhhhhhhhhhh <3 SeasonicMikeC wrote:An early pre-production sample of the 750W model but I am under NDA to stay mum. I've been told retail samples will be here by the end of the month. It is very impressive, clever technology and robust build.CA_Steve wrote:Mike - has Seasonic provided a sample, yet...or is one expected sometime soon?
Can't wait to plop a 750W into my Core i7 rig
-
- Posts: 2198
- Joined: Thu Feb 10, 2005 11:20 am
- Location: TN, USA
Hard OCP has first review. Thought it important enough to post this that I created account for first time after 3 years of reading SPCR but not allowed to post a link until get 3 posts on this board. Replace first two spaces with . and rest of spaces with /
www hardocp com article 2009 09 25 seasonic_x_series_x750_power_supply_review
MSRP: $199.99
www hardocp com article 2009 09 25 seasonic_x_series_x750_power_supply_review
MSRP: $199.99
-
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
Good piece.Pendan wrote:Hard OCP has first review. Thought it important enough to post this that I created account for first time after 3 years of reading SPCR but not allowed to post a link until get 3 posts on this board. Replace first two spaces with . and rest of spaces with /
www hardocp com article 2009 09 25 seasonic_x_series_x750_power_supply_review
MSRP: $199.99
I certainly like paul's reviews. He's no nonsense and generally to the point. But I'm also looking forward to your review Mike! Hope this will make us forget all the antec phantom and other semi-fanless designs!MikeC wrote:Good piece.Pendan wrote:Hard OCP has first review. Thought it important enough to post this that I created account for first time after 3 years of reading SPCR but not allowed to post a link until get 3 posts on this board. Replace first two spaces with . and rest of spaces with /
www hardocp com article 2009 09 25 seasonic_x_series_x750_power_supply_review
MSRP: $199.99
-
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
Since that review is up, I can mention a few things.merlin wrote:I certainly like paul's reviews. He's no nonsense and generally to the point. But I'm also looking forward to your review Mike! Hope this will make us forget all the antec phantom and other semi-fanless designs!
1) The fan does not turn on in the SPCR test/hot box till about 150W load, with ambient ~23-26C. When the load drops to below ~150W, the fan does stop again, but not instantly -- just depends on how quickly the heat is dissipated from not only the PSU but also the case.
2) The above means that it will be silent at idle or low load in most systems.
3) It really is very efficient.
Beyond that, it has great electrical performance. There are the fine details, but the gist is right there for the vast majority of silencers.
Would it be possible to fit this PSU inside NSK3480 without any modifications? I think the extra length would block the standard cable hole.
Edit: Nevermind, I bought the Nexus Value instead.
Edit: Nevermind, I bought the Nexus Value instead.
Last edited by sampo on Tue Sep 29, 2009 6:02 am, edited 1 time in total.
-
- Posts: 24
- Joined: Sun Apr 05, 2009 11:17 am
- Location: Washington, DC
I'm considering this PSU, but I'm concerned about SATA 3 compatability. The Seasonic warranty covers an excellent 5 years, but SATA 3 is due out within a year, if not sooner. I couldn't stomach paying $200+ dollars for a PSU to see it become obsolete in such a short time.
This new Seasonic is completely modular, yes, but if even they could make new SATA 3 modular cables after the fact, what are the chances of them being compatible with this unit?
This new Seasonic is completely modular, yes, but if even they could make new SATA 3 modular cables after the fact, what are the chances of them being compatible with this unit?
-
- Posts: 2198
- Joined: Thu Feb 10, 2005 11:20 am
- Location: TN, USA
I'm under the impression that the PSU side of SATA isn't changing in any way that would affect cabling.
http://en.wikipedia.org/wiki/Serial_ATA
http://en.wikipedia.org/wiki/Serial_ATA
Ya I haven't heard a single comment about changing the psu connector. It makes sense since it's already pretty small and if anything, most sata devices are reducing power requirements rather than increasing. SSD's will only get more popular over time also.dhanson865 wrote:I'm under the impression that the PSU side of SATA isn't changing in any way that would affect cabling.
http://en.wikipedia.org/wiki/Serial_ATA
Did you check for this behavior of going just getting better with load:MikeC wrote:Beyond that, it has great electrical performance. There are the fine details, but the gist is right there for the vast majority of silencers.
The DC Output Quality of the X-750 presents us with something of an interesting case as the results are very good but "backwards" from what we are used to seeing in a traditionally designed PSU... this unit actually started out at its highest value of 35mV and then dropped to ~10mV as it was more heavily loaded.
http://www.hardocp.com/article/2009/09/ ... y_review/8