Looks like Thermaltake silently nerfed their GF A3

[sLowEnd]

Thermaltake Toughpower GF A3 750W ATX v3.1 PSU Review - Silent Platform Change, Messy Protection Features & Fake 80 Plus Badge - Hardware Busters

Hardware Busters - Thermaltake Toughpower GF A3 750W ATX v3.1 PSU Review - Silent Platform Change, Messy Protection Features & Fake 80 Plus Badge - PSUs

hwbusters.com

 

[Bill_Bright]

Fake 80 Plus Badge

If true, shame on Thermaltake. I hope this has been reported to CLEAResult for their investigation.

I do believe the author's criticisms about changing OEMs requiring a different series name, as a general complaint, is a bit harsh, if not misguided.

It is not uncommon for manufacturers to use more than one factory to produce their products - even factories located in different countries. Hard drive makers do this all the time. So do car makers. This practice, besides increasing supply, eliminates the risks involved when having a "single source". These risks include factory fires, labor strikes, or even typhoons blowing the factory apart.

In these "general" scenarios, the typical, accepted method of product identification is to use the same series and model numbers, but with a different revision number. Or perhaps a very similar model number but with one letter changed or added to designate a specific factory.

HOWEVER - this practice also assumes the design and product specs are the same.

In this "specific" TT case, it appears there are significant design changes in the platform itself and therefore, IMO, at least the model number should be significantly different, making it obvious to consumers it is of a different design.

And absolutely for sure, any certifications (such as 80 PLUS) should be 100% legitimately earned for that "specific" design.

I say again, shame on Thermaltake.

 

[dirtyferret]

saw this, typically you see the brand/oem nerf the internal parts after the first run

And absolutely for sure, any certifications (such as 80 PLUS) should be 100% legitimately earned for that "specific" design.

one of the many reasons I've always said the 80 plus badge should never be used in terms of "quality" of the PSU

 

[FreedomEclipse]

saw this, typically you see the brand/oem nerf the internal parts after the first run

Some times its not about the nerfing but also component availability. This seems to happen a lot with tech in general. Components will differ over time because of availability. But so long as the end product is still within the stated performance, you cant point the finger and accuse them of being dodgy and scamming you. It would be hard to argue it in court if it went that far. A lot of manufacturers are now very careful what they say when it comes to advertising of products like when an ISP says they have speeds 'up to' xxxxmb/s etc.


Its when the manufacturers do a switcheroo and the same product is inferior to the ones that came before it is when there is a real issue. I think Kingston got caught doing this with some of their SSDs. Had a completely different product with the same model number sent out for reviews but switched it out with an inferior product come product release day. Now that is extremely scummy.

 

[dirtyferret]

Some times its not about the nerfing but also component availability. This seems to happen a lot with tech in general. Components will differ over time because of availability.

typically you have an approved part list from the brand with parts that are similar in quality but I've seen 105c caps replaced with 85c caps, ball bearing fans replaced with sleeve bearings, etc., it's more about lowering costs than simply changing suppliers with some of the brands/oems.

I think Kingston got caught doing this with some of their SSDs. Had a completely different product with the same model number sent out for reviews but switched it out with an inferior product come product release day. Now that is extremely scummy.

Adata got caught as well

 

[Bill_Bright]

one of the many reasons I've always said the 80 plus badge should never be used in terms of "quality" of the PSU

Ummm, nah! That is not a valid reason, regardless if the badge is genuine and rightfully earned, or fake, counterfeit, an unearned.

80 PLUS certification is not, and never has been an indication of "quality". Nor has the 80 PLUS certification program ever promoted it as such. It is simply an indication of efficiency - that is, at least 80% efficient at 20, 50 and 100% loads (and 10% load for Titanium). While it generally takes a better design and better parts to achieve these efficiency levels, nothing about quality, longevity, or reliability is part of the criteria for 80 PLUS certification.

So you are absolutely correct to say 80 PLUS is not an indication of quality, the possibility the 80 PLUS logo may be fake is not one of them.

Some times its not about the nerfing but also component availability. This seems to happen a lot with tech in general. Components will differ over time because of availability.

This is absolutely true. However...

But so long as the end product is still within the stated performance, you cant point the finger and accuse them of being dodgy and scamming you.

...this is where there's a problem with this specific TT PSU because they clearly changed the design too. It is NOT just about "performance" but advertised features too. In this TT case, they have different connectors, different noise ratings, different fan, and significantly, a different continuous-load maximum operating temperature.

Yet they keep the same model number. Setting aside the fake 80 PLUS certification, it may not be a "scam" (if the box accurately claims the applicable specs), it is at least "dodgy" to use the same model number - ESPECIALLY if the consumer does not know which model they will get when they place an order with Amazon.

 

[wNotyarD]

...this is where there's a problem with this specific TT PSU because they clearly changed the design too. It is NOT just about "performance" but advertised features too. In this TT case, they have different connectors, different noise ratings, different fan, and significantly, a different continuous-load maximum operating temperature.

Yet they keep the same model number. Setting aside the fake 80 PLUS certification, it may not be a "scam" (if the box accurately claims the applicable specs), it is at least "dodgy" to use the same model number - ESPECIALLY if the consumer does not know which model they will get when they place an order with Amazon.

This change of connectors has the potential for some serious issues for RMA's. Heck, the whole platform swap would make a RMA quite the loss.

 

[FreedomEclipse]

typically you have an approved part list from the brand with parts that are similar in quality but I've seen 105c caps replaced with 85c caps, ball bearing fans replaced with sleeve bearings, etc., it's more about lowering costs than simply changing suppliers with some of the brands/oems.

...this is where there's a problem with this specific TT PSU because they clearly changed the design too. It is NOT just about "performance" but advertised features too. In this TT case, they have different connectors, different noise ratings, different fan, and significantly, a different continuous-load maximum operating temperature.

This is absolutely scummy then and they should be called out on it.

 

[Bill_Bright]

This is absolutely scummy then and they should be called out on it.

No argument from me on that. And reported to CLEAResult about the 80 PLUS logo too.

 

[Ruru]

saw this, typically you see the brand/oem nerf the internal parts after the first run


one of the many reasons I've always said the 80 plus badge should never be used in terms of "quality" of the PSU

Exactly, after scoring on reviews, it's time to save on quality since customers think that they will be the same in the future. And I agree on the 80 Plus badge being a quality mark, I remember seeing Platinum units with C(r)apXon caps for example.

 

[Assimilator]

Thermalfake being scummy again, I see.

In this TT case, they have different connectors

Technically correct, but not materially relevant to most users - the 12+4 connector was changed to the 12V-2×6 one, which is technically an upgrade and is also backwards-compatible with existing cables.

 

[wNotyarD]

Technically correct, but not materially relevant to most users - the 12+4 connector was changed to the 12V-2×6 one, which is technically an upgrade and is also backwards-compatible with existing cables.

Only if the pinout on the PSU and cables wasn't changed - the layout was (pictures of the old and the new GF A3 below).

 

[Outback Bronze]

Hmm,

Seems far off their 1050w GF A3: Thermaltake Toughpower GF A3 1050W PSU Review - Hardware Busters (hwbusters.com)

"The Thermaltake Toughpower GF A3 1050W will be included in my best ATX v3.0 & PCIe 5.0 PSU picks article"

Was going to say if this is related to all of their GF A3 models but thankfully not.

 

[Vayra86]

typically you have an approved part list from the brand with parts that are similar in quality but I've seen 105c caps replaced with 85c caps, ball bearing fans replaced with sleeve bearings, etc., it's more about lowering costs than simply changing suppliers with some of the brands/oems.


Adata got caught as well

Lexar also appears to have differences between runs of NM790s. Jury is still out if any are inferior, they dont seem to be

 

[Assimilator]

Only if the pinout on the PSU and cables wasn't changed - the layout was (pictures of the old and the new GF A3 below).
View attachment 360260 View attachment 360261

The fact that the layout was changed is irrelevant, the pinout remains the same so the same cables will work fine with either unit. The labelling is clear and it's not possible to plug cables into the wrong connectors anyway, unless you're an idiot and force them.

 

[wNotyarD]

The fact that the layout was changed is irrelevant, the pinout remains the same so the same cables will work fine with either unit. The labelling is clear and it's not possible to plug cables into the wrong connectors anyway, unless you're an idiot and force them.

See, this is the part at least I don't have info on. Do you?
Do you remember the mess EVGA got itself into for different pinouts on a RMA'd supply which went through a project revision? And which did NOT change its layout nor its connectors, mind you.

 

[Ruru]

See, this is the part at least I don't have info on. Do you?
Do you remember the mess EVGA got itself into for different pinouts on a RMA'd supply which went through a project revision? And which did NOT change its layout nor its connectors, mind you.

It's not that long ago when I talked here about that EVGA PSU case, was it actually with you or with someone else? :E

 

[wNotyarD]

It's not that long ago when I talked here about that EVGA PSU case, was it actually with you or with someone else? :E

I do remember being part of such a discussion, yes. Especially because I did have an EVGA GQ supply, although mine was a lesser power model than the one which made the news and thus a different platform (the GQ's differ between those ≤750W and those ≥850W).

 

[Assimilator]

See, this is the part at least I don't have info on. Do you?

My assumption is that Aris tested this in his review and would have been very explicit if there was a compatibility issue.

Do you remember the mess EVGA got itself into for different pinouts on a RMA'd supply which went through a project revision? And which did NOT change its layout nor its connectors, mind you.

I actually hadn't heard of this, but it sounds like completely and utterly retarded behaviour from EVGA. Changing pinouts is a new product, not a revision.

 

[Ruru]

I actually hadn't heard of this, but it sounds like completely and utterly retarded behaviour from EVGA. Changing pinouts is a new product, not a revision.

Someone RMA'd his/her EVGA PSU, EVGA said that keep the modular cables, he/she got a "same" model from RMA, all storage fried thanks to changed pinout for the SATA cables.

 

[Assimilator]

Someone RMA'd his/her EVGA PSU, EVGA said that keep the modular cables, he/she got a "same" model from RMA, all storage fried thanks to changed pinout for the SATA cables.

That should be a lawsuit against EVGA.

 

[wNotyarD]

I actually hadn't heard of this, but it sounds like completely and utterly retarded behaviour from EVGA. Changing pinouts is a new product, not a revision.

Agreed.

My assumption is that Aris tested this in his review and would have been very explicit if there was a compatibility issue.

Aris at least says "The cables are the same between these two units". I must not have read it in the review earlier. The EVGA episode made me wary of such stealth platform changes, hence I pointed the possibility.

 

[Ruru]

@wNotyarD can't remember that did I say this when we talked about the EVGA case, but that made me think that what if someone buys sleeved cables for a specific unit and a later revision has the changed pinout issue? Damn.

At least I got my Cablemod cables with my G2 when I bought it used so no worries here.

 

[Chrispy_]

This is absolutely scummy then and they should be called out on it.

It's Thermaltake, they've proved countless times that they ARE absolutely scummy.

People who don't view Thermaltake as scum amuse me when they make the 'surprised Pikachu' face over the latest Thermaltake douchebag move because Thermaltake do it so often it's basically a surprise when they do behave acceptably these days.

 

[Bill_Bright]

Technically correct, but not materially relevant to most users - the 12+4 connector was changed to the 12V-2×6

My point is that it "IS" technically and materially relevant to the topic of this thread. This PSU has been "revised". Therefore, at the very least the model number of this PSU should indicate a different revision number.

If I went into my local computer shop and saw several boxes of these PSUs on the shelve, I would have no clue there could be differences between any of the same model number products. If I didn't know what was what, but there were revision numbers, I would pick a box with the newest number - exactly like I pick the carton of milk with the "freshest" use by date. If there was no indication there were differences, I would be totally in the dark.

Would it matter in the end? No clue because apparently TT failed to update the product specs on the box!!!!!

I actually hadn't heard of this, but it sounds like completely and utterly retarded behaviour from EVGA. Changing pinouts is a new product, not a revision.

New product? Nah! New revision number? Yes. Absolutely! The PSU itself didn't change.

But that was NOT the problem. The problem is much MUCH bigger and not specific to EVGA. The problem is, there is no industry standard for the PSU end of modular cables. So the manufacturers are left to come up with their own design and pinouts. Note I said "manufacturers" and not "brands". Since many "brands" outsource manufacturing to various OEMs, unless specified by the brand, it is those OEMs that design the PSU ends of those cables. This means even within the same brand, they could be different.

To make matter worse, should the brand change OEMs for a particular model, the cables could change and that is what apparently happened with this particular EVGA model - that is, they changed OEMs.

This is why, even within the same brand (and apparently model number too), you cannot mix and match cables. This way inventory management can be a real PITA if you have multiple computers, all with modular cables.

What happened is EVGA (or their OEM) changed the PSU end of one of their modular cables. The component end stayed the same and complied with ATX standards. After that (my bold underline added),

Someone RMA'd his/her EVGA PSU, EVGA said that keep the modular cables, he/she got a "same" model from RMA

But that "same" model had cables with different pinouts on the PSU end of the cables.

What should have happened is EVGA should have sent a new, compatible set of cables with the replacement PSU.

it sounds like completely and utterly retarded behaviour from EVGA

Assuming EVGA even knew the OEM changed the pinout.

Again, the problem is no industry standard for the PSU ends of modular cables. The member companies that establish the ATX Form Factor standards are the only ones who can fix this and prevent this from happening again. With no standard, it just invites proprietary solutions - never good for consumers.