Seasonic Prime Titanium 650W PSU Review

[bit_user]

I love Seasonic PSUs. Seasonic FTW!!!!

Definitely.

Now, can anyone answer whether the new PSU is compatible with the sleeved cables in current/previous modular Seasonic PSUs?

 

[iam2thecrowe]

I often wonder about these long term warranties. A warranty usually only covers manufacturing faults. Any manufacturer could say at the 10 year mark, its expected wear and tear or environmental issues if there is even a spec of dust inside.

 

[bit_user]

I often wonder about these long term warranties. A warranty usually only covers manufacturing faults. Any manufacturer could say at the 10 year mark, its expected wear and tear or environmental issues if there is even a spec of dust inside.

Not only that, but the number of people who would bother to submit a claim surely drops with time, which helps offset the cost of offering longer warranties.

I think it makes sense to use warranties as a good indication of the service life of a product. However, once you get above 5 years for anything PC-related, I start to have doubts about how much you can really read into it.

BTW, I'm using some 20-year-old hi fi gear that had a 10 year warranty. Still works great.

 

[MasterMace]

Am I the only one who got thrown for a loop when they were referred to as United Chemi-Con?

 

[turkey3_scratch]

There are two branches of Chemi-Con.

 

[cyanid]

I was about to get this one when I read about the inrush. with 230V in my country, I doubt my old home cable/breaker will agree with that kinda inrush.

 

[cyanid]

double post ...

 

[JackNaylorPE]

I totally understand the ooh and wow for Titanium units, especially ones like this. There is a direct mathematical connection between efficiency, noise, and lifespan. Assuming hypothetical units A and B to be identical, except for B being more efficient, the increased efficiency of B results in increased lifespan or the ability to maintain the same lifespan with a slower-spinning and quieter fan on B. It's up to the engineers how they want to balance all that out, and with Titanium units there is more flexibility. Less efficient units usually plain-out require faster fans.

Also, this unit has incredible performance and also incredible holdup times for its price. Contrary to the price/performance graphs, I really do think this unit is a great purchase, it's the most efficient power supply, one of the absolutely quietest power supplies, should last extremely long, and most likely has longer lifespan than gold units like the RMx and EVGA G2 that sell well on the market. Think of a place like India, how helpful a high holdup time like this may be for somebody who doesn't feel like dishing out a ton of rupees for a UPS; the holdup time along could really help the PC remain running with the unstable AC they have there.

All things being equal, I agree.... but the fact is, things are almost never equal. Efficiency does correlate as you say, but **not necessarily**. We have other means to measure noise, make judgements as to life span. etc. Here I am going to rely more on the tested decibel level than whether it's Titanium or Platinum. As for hold up times and performance, the data is there... why would I ignore that and instead say "Oh it's Titanium it must have good hold up times"... look at the data, not the marketing. As for the fan, I don't own a PSU that I can hear with my ear next to it and this includes bronze, Silver and Gold units.

There are obvious correlations between vehicle weight, number of cylinders, displacement and other factors so when i want to know which gets between mileage, should i read sll those stats and make my decision that way ... or should I just look at the actual real test measured mpg ? My 8 cyclinder Envoy is heavier than my sons 6 cylinder Xterra and it gets more than 20% better mileage.

Just because Kenyans train in high altitude and that gives them an advantage in long distance running, we just don't automatically hand the gold medal to the Kenyans at then Olympics. We wait till they run the race and the guy who actually displays the best performance "by the numbers" gets the gold medal.

So while "usually" might put us in a position to make a good guess, why rely on a guess when the actual data is readily available ? It's kind of foolish i think to look at the box and say "ooh it's titanium rated, it must be low noise, it must have low ripple, it must have great hold up times etc"

So yes, when we have the data right in front of us, it is irrelevant that "Less efficient units usually plain-out require faster fans" ... usually isn't always, and why rely on "usually" we can see actual and measured fan speeds and noise is by reading the information in the review that is the context of this thread. The "Green" label doesn't guarantee that, the data does. The green label ONLY guarantees you that you will save a miniscule and non-recoverable amount of money. The numbers are the numbers, they are right there in front of us so no need to consider what usually might be.

I have several CP-850's from 2009 (80+ rated) that are among the best PSUs I ever owned ... and among the best that jonnyguru ever reviewed. He couldn't even hear the fan when under full stress in his hotbox.
http://www.jonnyguru.com/modules.php?name=NDReviews&op=Story2&reid=142

 

[turkey3_scratch]

At Jonnyguru they can never hear PSU fans in the hotbox because the hotbox itself has a ton of fans that are extremely loud screamers; they can't hear crap there.

 

[JackNaylorPE]

I think it makes sense to use warranties as a good indication of the service life of a product. However, once you get above 5 years for anything PC-related, I start to have doubts about how much you can really read into it.]

Was a time that that was true but warrantees today are far to often simply "insurance policies". It's simply a numbers game which is why on just about anything you can extend the warranty by paying more.

The quality of a component does certainly affect how long it lasts. When setting a price at say $100, manufacturer will look at the calculated failure distribution of say

2% will fail in 1 year
3 % will fail in 2 years
... and so on till 10 years

and make a determination, that that model will "turn a profit" with a 3 year warranty at a cost of $100. That same exact model can still turn a profit with a 5 year warranty but they'd have to charge $115, to give 7 years, they might have to charge $135.

This was once common with HDs where the same drive was sold at one price with one warranty and at a higher price for the same drive with a different model name and longer warranty. Many consumer and enterprise drives differ only by firmware and yet the enterprise drives cost much more and have longer warrantees.

It was not long ago that the EVGA had a 600 or 650 watt B / B1 model that was being sold with a 10 year warranty. The higher quality B2 is 5 years, G2 7 years, P2 10 years. The B/B1 models available now are only 2 years

 

[JackNaylorPE]

At Jonnyguru they can never hear PSU fans in the hotbox because the hotbox itself has a ton of fans that are extremely loud screamers; they can't hear crap there.

read the review... Also In have built 4 boxes with them.... under full load in SLI overclocked CPU / GPUs full stress, there is no sound, even with side panel off. And the huge 120 mm fan is mounted externally where the cables plug in so any sound made would be bvious.

 

[turkey3_scratch]

I think it makes sense to use warranties as a good indication of the service life of a product. However, once you get above 5 years for anything PC-related, I start to have doubts about how much you can really read into it.]

Was a time that that was true but warrantees today are far to often simply "insurance policies". It's simply a numbers game which is why on just about anything you can extend the warranty by paying more.

The quality of a component does certainly affect how long it lasts. When setting a price at say $100, manufacturer will look at the calculated failure distribution of say

2% will fail in 1 year
3 % will fail in 2 years
... and so on till 10 years

and make a determination, that that model will "turn a profit" with a 3 year warranty at a cost of $100. That same exact model can still turn a profit with a 5 year warranty but they'd have to charge $115, to give 7 years, they might have to charge $135.

This was once common with HDs where the same drive was sold at one price with one warranty and at a higher price for the same drive with a different model name and longer warranty. Many consumer and enterprise drives differ only by firmware and yet the enterprise drives cost much more and have longer warrantees.

It was not long ago that the EVGA had a 600 or 650 watt B / B1 model that was being sold with a 10 year warranty. The higher quality B2 is 5 years, G2 7 years, P2 10 years. The B/B1 models available now are only 2 years

The EVGA NEX G does have a 10 year warranty. I don't know about the NEX B. But you are confusing the NEX 650B with the 600B. Totally different, and the 600B was always a 2 year warranty, never was 10. G2 is also 10 years except the 550W one.

At Jonnyguru they can never hear PSU fans in the hotbox because the hotbox itself has a ton of fans that are extremely loud screamers; they can't hear crap there.

read the review

I've read about every Jonnyguru review. If you read a lot, you'll realize that they always mention their screamer fans that the Sunmoon load tester has which makes it impossible to hear the majority of power supplies.

 

[JackNaylorPE]

The EVGA NEX G does have a 10 year warranty. I don't know about the NEX B. But you are confusing the NEX 650B with the 600B. Totally different, and the 600B was always a 2 year warranty, never was 10. G2 is also 10 years except the 550W one.

No absolutely no confusion ... I made about a dozen posts here on THG when it was happening where someone posted that they were thinking of buying them and getting a dozen responses saying how crappy it was. My response was "construction is what it is (ugh) but as long as EVGA is offering a 10 year warranty, it's not like if it goes south, you are out any money". It only lasted a few months and I don't know why they didn't ... perhaps they had a tom of them they couldn't move but there is no doubt that this was offered at one time.

I've read about every Jonnyguru review. If you read a lot, you'll realize that they always mention their screamer fans that the Sunmoon load tester has which makes it impossible to hear the majority of power supplies.

Obviously didn't read this one.

As always, the SunMoon SM-268 will be the Hall to my Oates, and... wait a minute here, I wanna be Hall. I'm being Hall and you can't stop me, Oates

Hot boxing this unit was fun. And by fun, I mean challenging. And by challenging, I mean insanely difficult. Friends, my hot box is an old ATX desktop case with a standard PS/2 sized power supply mount located directly above the motherboard I/O panel opening. I had to stand this thing on its side inside the case and point it at both the PSU opening and the I/O opening. Even then, the CP-850 nearly came up high enough to keep me from putting the top panel back on. As a result, the hot box rose to insane temperatures, up to 55 degrees. But if you look at the above numbers, the CP-850 didn't seem to care. It just shrugged and said, "whatever, man, this is nothing to me."

As usual, when the heat came on the efficiency came down a bit, and the rails came down a bit, but that was about it. I couldn't even hear that 120mm fan, that's how dull the hot box phase was. Looks like Antec's vision of a more efficient cooling system works.

Oh, good golly. Am I in heaven? I was almost sure this unit wasn't a Signature anymore from the load test numbers, but would you look at these scope shots for a second? Not only did I have to zoom the scope in to even see what little ripple there was, I had to struggle to not jump around and yelp with glee. I even turned on the cursors to show you how awesome the view was. Seriously, read this number and tell me you're not excited: 11.2mV at 12V, test five. Did I just hear some of you faint? Sorry... I did too the first time I saw that. That's actually better than my Signature 850W did. Same with the other two rails.

Seriously. This thing is so overbuilt, it's scary. Who do they have working at Delta anyway, Tim "The Toolman" Taylor? Doc Brown perhaps? No wonder ripple was so low - these parts weren't even breathing hard at full power.

Here he is raving about the cooling efficiency and how dull the hot box phase was and that he **couldn't even hear the fan**. So answer me this... why is he raving about that fact if, as you claim, .... it's an every day thing ?

And lets not forget, efficient cooling is not the only hi-end feature of the design which is the subject at hand. You've jumped on this one thing at the exclusion of all else but he's raving about everything... the reviewer clearly was of the opinion that the fact that he couldn't hear the fan was "unusual" to say the least. But what about all the other performance criteria which caused the reviewer to think he was in heaven and jump around with glee ? Not exactly a common occurence on this site.

This being a mere 80+ rated unit obtaining a performance level to make wolfie dance with glee should simply not be possible if we are to accept the postulation that it is necessary to have super high platinum / titanium efficiency to accomplish any of these performance goals.

Let's go to another source ....

SilentPCReview,com lists the CP-850 on its list of Editor's Choice power supplies noting:

Products ranked as Editor's Choice are "best in class" for the various parameters we consider most important: Low noise, high efficiency, high build quality, good value, useful features (more or less in that order)

An innovative PSU with 120mm in-line fan .... the CP-850 is just about as quiet as the Nexus Value 430 at lower power, stays extremely quiet and cool in those cases to >600W, with exactingly clean, regulated voltages lines at a price that's a pittance for such quiet, abundant, performance.

The above test data makes the CP-850 the quietest PSU ever tested

So clearly, titanium ratings are by no means necessary to build a quiet PSU.... more importantly, titanium ratings are by no means necessary to build a high performance PSU, this one is rated at only 80+ and, without question, excels in both.

 

[turkey3_scratch]

So clearly, titanium ratings are by no means necessary to build a quiet PSU....

Nobody made that claim.

 

[JackNaylorPE]

There is a direct mathematical connection between efficiency, noise, and lifespan.... the increased efficiency of B results in increased lifespan or the ability to maintain the same lifespan with a slower-spinning and quieter fan

No there .... obviously... isn't. The CP-850 was the "quietest PSU ever tested" and has terrible efficiency

 

[turkey3_scratch]

There is a direct mathematical connection between efficiency, noise, and lifespan.

No there .... obviously... isn't

Yeah there is. Going back to your Antec PSU, if it was more efficient, it would last longer, and could even have a more relaxed profile than it already has. That follows the mathematical connection. But nobody made the claim that "titanium ratings are necessary to build a quiet PSU".

 

[JackNaylorPE]

Apparently you are now in violent agreement with exactly what I said above with regard to efficiency ..... Glad to see you come around

Efficiency does correlate as you say, but **not necessarily**. We have other means to measure noise, make judgements as to life span. etc. Here I am going to rely more on the tested decibel level than whether it's Titanium or Platinum. As for hold up times and performance, the data is there... why would I ignore that and instead say "Oh it's Titanium it must have good hold up times"... look at the data, not the marketing. As for the fan, I don't own a PSU that I can hear with my ear next to it and this includes bronze, Silver and Gold units.

So while "usually" might put us in a position to make a good guess, why rely on a guess when the actual data is readily available ? It's kind of foolish i think to look at the box and say "ooh it's titanium rated, it must be low noise, it must have low ripple, it must have great hold up times etc"

So yes, when we have the data right in front of us, it is irrelevant that "Less efficient units usually plain-out require faster fans" ... usually isn't always, and why rely on "usually" we can see actual and measured fan speeds and noise is by reading the information in the review that is the context of this thread. The "Green" label doesn't guarantee that, the data does. The green label ONLY guarantees you that you will save a miniscule and non-recoverable amount of money. The numbers are the numbers, they are right there in front of us so no need to consider what usually might be.

As for lasting longer ... no,not necessarily for example, what if tpo met a price point, they bought high efficiency electronics but cheaped out on the fan bearing and fan dies after 2 years ?

I'll assume there's also no longer any disagreement on the 10 year warranty and the fact the JG thot not hearing the fan was a big deal.

 

[turkey3_scratch]

You rely on tested decibel level which Jonnyguru does not do. Nobody said anything about a Titanium unit having good holdup time. Everything you say in that quote is nothing I said. I was speaking of a balance, and you are referring to me saying Titanium units always have low noise and always last long. Nope I didn't say that, so you just are misinterpreting what I said.

 

[JackNaylorPE]

Nobody said anything about a Titanium unit having good holdup time.

The "I didn't say what we can quote from the posts above" thing is becoming a recurring theme Was it nobody who posted this on September 7, 2016 4:23:49 PM ? Tho you didn't say "good", you said "incredible" and 'high".

Also, this unit has incredible performance and also incredible holdup times for its price....Think of a place like India, how helpful a high holdup time like this may be for somebody who doesn't feel like dishing out a ton of rupees for a UPS; the holdup time along could really help the PC remain running with the unstable AC they have there.

Everything you say in that quote is nothing I said.

Everything I have said is quoted from your posts starting with your introductory statement

I totally understand the ooh and wow for Titanium units., There is a direct mathematical connection between efficiency, noise, and lifespan. Assuming hypothetical units A and B to be identical, except for B being more efficient, the increased efficiency of B results in increased lifespan or the ability to maintain the same lifespan with a slower-spinning and quieter fan on B

I still haven't seen the justification for the ooh and wow part. Like the hold up time statement that you "didn't say", there's no ambiguity in your post here on the board, It doesn't say **might** result in increased lifespan ... it doesn't say **might** allow the same lifespan with a quieter fan.

When you wrote "Assuming hypothetical units A and B to be identical".... I very clearly prefaced my responses on a very important distinction:

All things being equal, I agree.... but the fact is, things are almost never equal.

Yes, so what we have here is what we agree is true in the theoretical instance versus the reality of the marketplace. The review initiating this thread and the position I have put forth is based upon the assumption that the discussion context is units like the Seasonic Prime available in the real world marketplace. And when we limit the discussion to what one can actually buy, there is no 100% guarantee that a titanium rating brings anything to the table other than an implied cost of electricity. When you see the titanium rating, you may take (assuming the manufacturer isn't lying and it holds up in testing) that the unit will deliver the following:

at 10% load the efficiency is 90% or better
at 20% load the efficiency is 92% or better
at 50% load the efficiency is 94% or better
at 100% load the efficiency is 90% or better

That's it, nothing more. Anything beyond that can only be characterized as possible, maybe even likely or highly probable but the assumption of it being 100% true is certainly not something that I think one should rely on for purchasing decisions, especially in this price range. Especially when there is no need to imfer an iffy conclusion from an efficiency rating when verified actual test data on the various performance criteria you believe the rating implies is readily available.

As per above .... Kenyan runners have the routine opportunity to run in atmospheric conditions which give them a significant competitive advantage in long distance running. Paraphrasing your words ... "all things being equal", as the atmospheric conditions in Kenya produce better lung capacity and oxygen transfer efficiency in the blood, Kenyan athletes will be able to maintain peak running speeds with slower breathing and the endurance advantage **results** in Kenyans winning all the races. Again, the problem is that it doesn't say **may result** and all things are not equal .... nutrition, training methods, use of hyperbaric chambers, sheer will, stride length, individual biology all factor in to who wins the race.

I have agreed with you from the getgo that higher efficiency is an advantage and that **all other things being equal**, this does carry over and beneficially impact other performance areas. My reservation is we don't have access to a marketplace where **all other things are equal** and since we don't, the efficiency rating is a guarantee of nothing other than relative electricity costs. These carry over benefits may be offset by other criteria.

Again I am using fans as it is a simple concept to use as an example rather than go thru every test result. But the same applies to every other performance criteria. Yes the lower efficiency / lower heat does give the unit an advantage ... but there are ways of offsetting those advantages as was evident in the CP-850 design which didn't even manage a bronze rating. The same PSU that SilentPCreview called the "quietest PSU they ever tested".

I really don't understand the relevance of the "jonnyguru doesn't test decibel level" comment. Other sites do, and ... in the context of this thread which is linked to a THG article which did in fact list and graph noise levels. Noise is an issue of concern for me.... I invest considerable sums in water cooling not for increased performance as with current CPUs / nVidia GPus, it's most often not attainable over air, I do it for decreased noise. The only fan you can hear in my son's water cooled (CPU only) w/ SLI'd air cooled 970s (w/ 9 case / rad fans) is the fan on the EVGA G2 1000. So yes, thius is something I have high on priority list so do go searching for this data when it's available.

If you are not saying that "Titanium units always have low noise and always last long"... or "always have good hold up times", great voltage stability, low ripple, and so on, then I gotta go back to the original statement.

I totally understand the ooh and wow for Titanium units, especially ones like this.

What's the ooh and wow factor of this Titanium ratings if this rating is not **always** a reliable indicator of any of those things ? If you are not saying that, then the Titanium rating does not necessarily mean that they will:

-have low noise
-last long
-have great voltage stability
-have great hold up times
-have low ripple
-have great voltage stability

If we can't depend upon it **always** than I don't see the ooh / wow factor unless of course the cost of electricity is a major decision factor. I don't even look at the "metal" rating... tho I live in an area of high electricity costs as compared to most of NA, I would consider it if I lived in many parts of Europe. But when we go PSU shopping should we:

a) See the Titanium rating, say "ooh wow" and stop looking ..?

b) or go thru the available review / test data ignoring what high efficiency might imply if **all things were equal** and see what readily available test result are for temps and noise (at various loads), ripple, voltage stability, inrush current, hot box performance, actual efficiency (we have all seen units not manage to pass what their ratings imply). Do we ignore protection features, infrared image results, transient results, build quality (i.e. does sloppy soldering and other poor workmanship matter) ?

This is not very different from the argument that the PC component that costs more, **all things being equal**, will be better as Titanium units *usually* cost more than Gold which in turn cost more than Silver and so on. Yes it does cost more to build better quality products and perhaps more often than not, a higher cost component will outperform another comparable all other things being equal lower cost unit but this is not always true ... and yet how often do we see peeps post a proposed build where they just picked the highest cost component in each pcpartpicker category ?

What I am trying to convey is that I am not misinterpreting what you said, after all I'm quoting your posts. I agree with you 100% on the **all things being equal part**. What I am saying is the "all things being equal" premise is not relevant to a purchasing decision as all things are not equal.

If you are not saying that "Titanium units always have / provide [insert anything here other than electrical efficiencies]", then w/o the **always**, it don't see how i could rely on it as a 100% reliable indicator of anything upon which to make a purchasing decision. What the Titanium rating may imply with regard to any measure of performance is not as reliable an indicator as the actual measured performance. I'll leave you with this

My wife's Italian ancestry (both sides) implies that all things being equal she should be the better option to cook (and can) the tomato sauce every other weekend given the exposure to a large family where weekend meals were attended by lots of relatives with many recipe ideas ... but it's the dopey half English / half Irish guy she was foolish enough to marry who performs the task. Wifie can be a very good cook, but she tends to multi-task while cooking (laundry, phone whatever). If we let Mom do the pasta cooking, no one here would have ever experienced "Al dente" ... after a taste test of "too hard" she'll tackle another task and when she comes back, Al dente left the building. She has taken over canning duties tho. She has to cook all the meat tho ... which she does very well.

It would seem that we agree on what's what under each scenario ... what we don't seem to agree on is what premise is applicable to decision making. If I understand correctly the rating appeals to you because it often does mean other advantages in other performance categories, I just don't get why the "metal" rating is of significant "ooh wow'" value when the actual data on those performance categories is readily available and is a more accurate indicator of where the unit stands in that regard.

 

[turkey3_scratch]

I never said the holdup time was good because of the efficiency; it is good because of the design of the power supply and the bulk caps used. Again you're putting words into my mouth.

 

[Virtual_Singularity]

While I understand the numbers, concepts, and more or less recognize what equates to quality in a PSU, I'm no expert by any means, and so have a couple questions. 1. Though I tend to think the unit's protections already cover this, even for areas where 240v is the standard. But nonetheless, do those who've cited concerns re Inrush Current with this unit have any reason to be worried whatsoever?

And, 2. While this PSU doesn't use capacitors in it's cables, there are some other PSUs that have proven reliability, among the longest warranties, and excellent reputations, that do. Is not using capacitors in cables merely a subjective preference, or is there actually a reason using such cables should be avoided by a manufacturer?

 

[turkey3_scratch]

While I understand the numbers, concepts, and more or less recognize what equates to quality in a PSU, I'm no expert by any means, and so have a couple questions. 1. Though I tend to think the unit's protections already cover this, even for areas where 240v is the standard. But nonetheless, do those who've cited concerns re Inrush Current with this unit have any reason to be worried whatsoever?

And, 2. While this PSU doesn't use capacitors in it's cables, there are some other PSUs that have proven reliability, among the longest warranties, and excellent reputations, that do. Is not using capacitors in cables merely a subjective preference, or is there actually a reason using such cables should be avoided by a manufacturer?

The inrush current is not that bad at all, and it shouldn't matter at all. Matter of fact, for the holdup time this unit achieves, inrush current is incredibly low.

Capacitors in the cables are used to lower ripple. It's preferred to not have them because they make the cables bulky; it's much better if a unit achieves low ripple without using cable capacitors but by just using the ones inside the PSU itself.

 

[Virtual_Singularity]

Thanks, understood and much appreciated.

 

[DuleGURU]

Seasonic Prime is the best pc power supply ever.