Does Gigabyte's "All Solid Caps" means made in Japan Solid State capacitors?

Happy Hopping

Happy Hopping

Supreme [H]ardness
Joined
Jul 1, 2004
Messages
7,837
http://www.gigabyte.us/products/product-page.aspx?pid=4567#ov

I don't know if solid state capacitors are ALL made in Japan or what the deal is.

I used to install this guy from gigabyte:

http://www.gigabyte.us/products/product-page.aspx?pid=4315#ov

as you can see, it clearly says "All Japanese Capacitors"

but rumor has it that this motherboard is being discontinue as Intel is killing the G2030 CPU.

So Gigabyte is asking people to switch to the B85 (first link), in which it DOES NOT state "all japanese capacitors" on the motherboard. Rather, on the website, it says "GIGABYTE uses All Solid Caps (Capacitors) and Low RDS(on) MOSFETs which are rated to work at higher temperatures to provide a longer lifespan."

but is it made in Japan though? If not, where else made Solid Caps?
 
Yeah, recently Gigabyte started cutting costs, avoiding japanese caps.
The caps in the first link, seem to be the same taiwanese caps that ASUS uses, Apaq.
The second link are SANYO Os-Con, which are japanese.
 
Would be cool to see a large comparison image with origin/manufacture next to the image of the CAP.
 
Not all solid/polymer caps are Japanese, so I'd say it's likely that at least the lower-end boards use some bad to mediocre brand like Teapo.
 
how many yr. different in lifespan between:

Japan solid state vs. Taiwan solid state

Japan solid state vs. china liquid capacitors

Also, I just got a confirmation from distribution that a no. of Gigabyte motherboard w/ Japan capacitor is being phrase out w/ solid state capacitor that is made in ??

So I don't know if I should re-post as a new topic: what brand name motherboard will still be using japanese capacitors?
 
Happy Hopping said:
So I don't know if I should re-post as a new topic: what brand name motherboard will still be using japanese capacitors?
Click to expand...

It's probably more that higher-end boards will use Japanese caps and lower-end boards will use Taiwanese or others. Not so much by brand. In any case if you're worried about it, just wait for reviews of a board that mention the capacitor brand(s) and/or provide high-res pictures that include the capacitors.

A Taiwanese solid electrolytic capacitor is highly unlikely to be as good as a Japanese one but at least they shouldn't be as bad as the old caps that failed like crazy on Abit and other boards.
 
Gigabyte gives 3 yr. warranty on all their motherboard, including the B85 series that c/w Taiwanese solid state capacitors, which replace the japanese capacitor.

Since it's still solid state capacitors, they should last at least 3 yr., would they?
 
I would say it's very unlikely for modern capacitors on a decent brand of motherboard to fail within 3 years. Far less likely than the old wet electrolytics that used to fail. Not impossible, but not a huge concern.

If you want to keep the thing running for 20 years, or perhaps if you intend to stress the board with things like heavy overclocking, I'd still consider Japanese capacitors an absolute requirement, personally.
 
No, all these motherboards that I will setup is for general household use, they use it mostly for internet. No over clocking. And once the 3 yr. is up, they don't care.

So as long as Taiwanese Solid State capacitors can last 3 yr. or more. It's good enough for me.

But I'm curious to know how many more year those japanese capacitors will last?
 
I don't think there's enough data on them yet to say. Good ones should last decades, but I bet even Taiwanese solid caps should last a good 10 years or more in non-high-stress environment (e.g. they aren't handling an excessive amount of ripple or getting too hot - caps shouldn't be placed right next to heat sources or should be insulated from them as much as possible). That is NOT based on studies so don't take it as an absolute truth.

But hey, even when they are next to heat sources, at least solid electrolyte isn't particularly vulnerable to the electrolyte drying out as it does in wet electrolytics. They can still be damaged by heat but data so far shows them to be less susceptible to that.

But we won't know about the long-term reliability of them for several more years, IMO.
 
Last edited:
Happy Hopping said:
Since it's still solid state capacitors, they should last at least 3 yr., would they?
Click to expand...

Observe this. This is a typical life solid polymer capacitor, japanese made (Chemicon)

9hm1ia.png


Observe also this. This is a typical taiwanese Apaq capacitor.

260ttav.png


Also this is typical Nichicon range:
http://www.digikey.com/us/en/ph/nichicon/alumcondpolymer.html

They all have rating 2000h@ 105C (except for few models at 1000h). BUT, as it says:

"Since the polymer is a solid, it also gives longer life, not following the classic Arrhenius formula. Instead of doubling life with every 10°C drop in temperature, the lifetime will be 10 times longer for every 20°C decrease in temperature."

So, theoretically:
- 2000h @ 105C
- 20000h @85C (2,2 years)
- 200000h @ 65C (22 years) <--- If your motherboard's capacitors are above 65C you either have serious cooling issues or some crazy high overclock.

There are also "long life" solid polymer capacitors, of 5000h, 8000h and 10.000h, but one needs to read the specific mark on the capacitor to know it and it's not in your ordinary motherboard.

So what's the big deal if one puts a japanese or a taiwanese or a chinese, if they all have same rating? The devil is in the details...

The difference lies in a small phrase that appears only in the Chemicon sheet:
"Failure raite: 0,5% per 1000h maximum (confidence level 60% at 105C).

So, despite the rating, there is a 0.5% chance of failure within 1000h of operation, with a confidence level of 60%. What does the confidence level mean? It's a term used in statistics. http://en.wikipedia.org/wiki/Confidence_interval
In poor words, it says how much certain we are, of what we stated as a statistical result. Meaning, that there is 0.5% chance of failure within 1000h and we are 60% certain of that statement to be true. 60% level of confidence, is pretty low. In some sciences, it's unacceptable (medicine usually needs 95% or above).

So, the problem is once more, the quality. The famous electrolytic capacitors that were exploding, were rated just as well, as the high quality caps that never exploded. The difference, was exactly, in the quality.

The 4 japanese brands, have a good track record for high quality, this is why some manufacturers were using them as pubblicity points. The taiwanese Apaq, is known to be cheaper brand, but since solid polymers are "new", there are not enough data to judge what's the quality gap and they do not show detailed statistical data like Chemicon does.

Bottom line, at the end, you either trust one company or you don't. Because, even if Chemicon says "x% chance of failure with Z confidence interval", this is an internal test made by Chemicon, and nobody external goes there to verify.

So it all goes down to how much confidence, the customer puts on their brand...
Certainly, some premature motherboard deaths that happen without apparent reason, are probably caused by some "early" capacitor failure, of the likes described by Chemicon. The solid polymers die "silently", so it is hard for the user to verify the cause of death, so we don't get many reports.

So, the life rating is one thing, but there is also the quality, which dictates, what are the chances of untimely demise, long before the "rated life" comes. For instance, is it better a capacitor rated 2000h with 0.5% failure per 1000h or a capacitor rated 5000h, with 2% failure per 1000h? I 'd pick the first one... Basically, one this is the rating. Another, equally important thing, is how reliable this rating is. Because a capacitor that has high chance of premature death for example, if it does fail, gives very little consolation if it was rated for 10000h. It's more important to be able to actually survive up to 2000h, rather than have a rating of 10000h and never arrive there. :)

This said, the current situation is much better compared to the electrolytic capacitors.
 
Last edited:
Ryujin Jakka said:
So if they started using taiwanese caps then which company is still using jap caps?
Click to expand...

One should find the model that interests him and try to find a zoomed in picture and see (better if the markings are readable). Gigabyte, is certainly dropping the japanese ones at least on cheaper boards (the Gigabyte UD3P that i have for example), i don't know what happens with the more expensive.

ASUS was the first to abbandon boat. If i remember correctly, on more expensive boards, ASUS puts longer life, but always taiwanese.

My guess is Asrock and MSI could be still using them (i know for sure that Asrock has been using Nichicon on even the low models) and i remember MSI was using Chemicon. But one must investigate a bit on his own, because GIgabyte changed, so anyone can change.
 
I have a MA-790x that's been running solid since '09 without a hitch.

Last build I used a GA-Z87 and it shit out in 9 months. RMA is in now, hopefully it works out ok.
 
A wholesaler told me this morning that solid capacitor from Taiwan vs. Japanese, whether it last 2000 hr. or not, is how well it absorb voltage surge from lightning.

So say a good capacitor would last 2000 hr., with voltage surge, it could drop to say 1800 hr. and if you can a handful of those lightning surge in the summer, you could kill the 2000 hr. down to 1000 hr.

I remember there are a few service phone calls every summer just after a thunder storm.

On unrelated matter, Asus just announced that 12K hr. capacitor, but I don't see the word "Japanese" next to it

http://www.asus.com/Commercial_Servers_Workstations/Z10PED8_WS/
 
Happy Hopping said:
A wholesaler told me this morning that solid capacitor from Taiwan vs. Japanese, whether it last 2000 hr. or not, is how well it absorb voltage surge from lightning.

So say a good capacitor would last 2000 hr., with voltage surge, it could drop to say 1800 hr. and if you can a handful of those lightning surge in the summer, you could kill the 2000 hr. down to 1000 hr.

I remember there are a few service phone calls every summer just after a thunder storm.

On unrelated matter, Asus just announced that 12K hr. capacitor, but I don't see the word "Japanese" next to it

http://www.asus.com/Commercial_Servers_Workstations/Z10PED8_WS/
Click to expand...

Without being a thunder expert, i think life and surge are unrelated. Chemicon has different mentions about surge tolerance and failure rate, implying that failure rate is related to manufacturing process (they can't have almost the same quality on every single capacitor, there is also random failure at play). Capacitors in a motherboard, deal with low voltages (see APAQ paper above). Thunders deal in thousands of volts. It's i think funny to expect a capacitor to withstand the voltage of a thunder. That's the business of the socket you connect the computer (there are anti-surge sockets) or UPS or as final line of defence, the PSU itself, that has MOV and anti-surge protection. The capacitor's surge is to deal with oscillations of the voltage given by the PSU, which should be within tolerable limits. In other words, if the PSU fails to stop the lightning, i think it's hilarious to expect a capacitor on the motherboard to do so...

On the other hand, temperature relates with life, as is seen in all ratings.

ASUS uses only Apaq for many years now. If you google "ASUS Apaq", you will find results that go back to 2007, as well as some photos from actually blown caps (unusual, but can happen).
 
For the history, APAQ is very easy to identify, because the capacitor is marked with a small logo (like a crescent with a dot):

2m84i9k.jpg


Nowdays, they follow that color scheme as far as i know, in the past they had other colours too (i don't remember whether red or blue), but they always have that logo. My Gigabyte 970 UD3P has exactly this grey scheme with the logo and it's the same that i 've casually observed on some ASUS boards.

Some companies use "one colour" (like all black or gold) caps for some motherboards models, but they 're the same manufacturer they use in their "regular" boards, just coated or painted to match the motherboard colour.
 
Happy Hopping said:
So I don't know if I should re-post as a new topic: what brand name motherboard will still be using japanese capacitors?
Click to expand...

Of the four largest brands, it's "all of them" or "none of them" depending on which board you buy. There's no consistent rule of thumb, and quite likely, none has ever existed for an extended period of time. I suppose this makes sense because every manufacturer faces the same market conditions, changes in material costs, and downward price pressures, hence the reliance on similar strategies to compensate.

Some observations:
  • High end boards generally have Japanese capacitors. We're probably looking at the $200+ range for a guarantee, but the exact cutoff point varies. It's possibly dependent on such factors as component costs, price pressures, and shipment expectations for the year.
  • Consequently, there's no perpetual pattern attributable to a particular manufacturer. A brand might economize on components one year and then rebound the next. In the long run, it's probably unlikely for the top dogs to stray wildly (either direction) from what the others offer.
  • Virtually everyone of note claims to stress durability, whether it's "Military Class", "5X Protection", "Durathon", or "Ultra Durable".
  • Specific marketing claims mainly apply to the exact time period during which they are made. For example, Gigabyte used solid capacitors on just about everything during their initial "Ultra Durable" marketing push, but they soon adopted Taiwanese caps on lower cost boards (i.e. the A55M-DS2 from 2011 is one such example). The latter is universal among their competitors as well.
  • Another strategy is to launch a model with higher quality components and then swap them with lower cost equivalents in later shipments (which may be paired with a price drop - they have to cut down somewhere).
  • One technician I spoke to was less interested in capacitor origin than the MOSFETs, chokes, and phase design. Supposedly, polymer caps are less likely to be the first source of failure, plus they can be easily replaced if you absolute insist on saving the board.

I've installed a few recent motherboards for friends, and I can confirm that the Gigabyte Z97X-UD3H (Nippon Chemi-Con) and ASUS Maximus VI Hero (Nichicon's FPCAP) boards have Japanese caps. The ASUS Z87-A is mostly FPCAP (5K) with a few APAQ APCON caps in less critical places. Gigabyte's AM1M-S2H has a mixture of APCON solid caps and older electrolytic caps (maybe Chemi-Con?).

Like ASUS, ASRock tends to specify FPCAP for more expensive parts. On lower cost parts, I saw APCON on a Pro3 board, while the H97 Anniversary uses something else that I couldn't narrow down.

Long story short, you have to research each board individually, buy higher priced models, and take the marketing/consumer myths (which aren't necessarily lies, but they're short-term observations repeated as long-term truths) with a grain of salt. There's no superhero or evil villain in this challenging business.
 
You must log in or register to reply here.
Back
Top