Channel Well Technology CSX 850M-G

[crmaris]

The first ATX 3.0 compatible SFX power supply falls into our hands, and we don't miss the opportunity to evaluate it thoroughly. The Channel Well Technology CSX 850M-G offers high performance, coming with a single 12+4 pin PCIe connector.

Show full review

 

[Dragokar]

Thanks for the preview about upcoming Corsair SFX ATX3.0 modells. I do hope that they still tweak some parts.

 

[mama]

We need more power supply reviews on ATX 3.0 products (when they are released).

 

[Xeanoa]

TPU, where is your proof reader?
This article needs some love. SI units are all over the place

 

[R-T-B]

SI units are all over the place

Pretty sure that's intentional. No one uses F temps in PC space, and for scientific articles in general, regardless of nationality, SI is the norm.

 

[Mussels]

It's a good sneak peak at whats to come from brand name PSU's soon, disapointing that it has some flaws on 240V

 

[Xeanoa]

Pretty sure that's intentional. No one uses F temps in PC space, and for scientific articles in general, regardless of nationality, SI is the norm.

I don't expect anybody to use F either.
But is it 10°C, or 10 °C? The article doesn't seem to be sure about it, and neither about V, W or %.

 

[PCL]

I thought a 12+4 connector wasn't required for ATX 3.0 certification. Are we sure that all ATX 3.0 power supplies would be PCIe 5.0-ready?

Edit: maybe I was misunderstanding this article?

ATX 3.0 PSU Specification Loophole Lets Manufacturers Evade Stringent Excursion Tolerance Testing by Simply Excluding the 12+4 pin Connector

When testing recent high-end GPU launches such as the AMD Radeon RX 6950 XT graphics card, we observed some system reboots that caused due to power-draw spikes (technically known as "excursions"). This is when a graphics card, for a brief moment, draws more power than the connector is capable of...

www.techpowerup.com

 

[crmaris]

I don't expect anybody to use F either.
But is it 10°C, or 10 °C? The article doesn't seem to be sure about it, and neither about V, W or %.

the tables don't have space because they are automatically generated. You are new here. You will learn more about PSU reviews, soon

I thought a 12+4 connector wasn't required for ATX 3.0 certification. Are we sure that all ATX 3.0 power supplies would be PCIe 5.0-ready?

Edit: maybe I was misunderstanding this article?

ATX 3.0 PSU Specification Loophole Lets Manufacturers Evade Stringent Excursion Tolerance Testing by Simply Excluding the 12+4 pin Connector

When testing recent high-end GPU launches such as the AMD Radeon RX 6950 XT graphics card, we observed some system reboots that caused due to power-draw spikes (technically known as "excursions"). This is when a graphics card, for a brief moment, draws more power than the connector is capable of...

www.techpowerup.com

it is not required as it seems, this is a new thing, and we are waiting for Intel actually to confirm this (which it will do soon). Till the ATX 3.0 spec is officially changed, though, we keep to what we have.

It's a good sneak peak at whats to come from brand name PSU's soon, disapointing that it has some flaws on 240V

many PSUs have flaws with 230V, especially in PF, because nobody has been certifying them with that voltage so far. 80 PLUS 230V desktop units are too few. This changed with Cybenetics which also tests with 230V ALL units.

 

[v12dock]

Another top tier review from crmaris. Seriously I greatly appreciated your reviews.

 

[fb020997]

Thanks for the preview about upcoming Corsair SFX ATX3.0 modells. I do hope that they still tweak some parts.

Full Japanese caps, a bit of tuning on efficiency/PF and a different metal shell, and you have the Corsair SF850

 

[mechtech]

Nice thorough review as always.

A possible suggestion............to help out the layman.........

Perhaps beside the "Hold up Time" add "ATX Spec 3.0 xx ms max" or whatever etc. etc. for each graph.

Might help give people an idea of how it measures up to the specification.

 

[shovenose]

I don’t see how you can “highly recommend” a product with Elite and ChengX caps. Sorry.

 

[Ruru]

I don’t see how you can “highly recommend” a product with Elite and ChengX caps. Sorry.

Agree here. Teapo is the only non-Japanese cap manufacturer I'm fine with.

 

[LuxZg]

I know it's been avoided for a reason, but could we get approximate price range? I'd love a 650W version of this model, hopefully retaining 300W 12+4 connector. News about more humane models (650-750W) of all these ATX 3.0 PSUs are severely missing

 

[shovenose]

Agree here. Teapo is the only non-Japanese cap manufacturer I'm fine with.

Agreed. I have no issues with Teapo in a well designed PSU.

 

[Assimilator]

Weird how the first true ATX 3.0 PSU you get your hands on is an SFX model, not standard ATX - but thanks for the preview.

I know it's been avoided for a reason, but could we get approximate price range? I'd love a 650W version of this model, hopefully retaining 300W 12+4 connector. News about more humane models (650-750W) of all these ATX 3.0 PSUs are severely missing

How is CWT supposed to know what the OEMs are going to price their products at, FFS?

 

[Zareek]

ATX 3.0, does that mean we are kicking the adoption of ATX12VO down the road a bit more? Not that I'm complaining...

 

[Assimilator]

ATX 3.0, does that mean we are kicking the adoption of ATX12VO down the road a bit more? Not that I'm complaining...

ATX12VO was DOA because there are still too many parts of the PC ecosystem that use 3.3V and 5V - for example PCIe (and therefore M.2 NVMe) slots use the former and SATA the latter - and motherboard manufacturers weren't willing to have the cost of rectifying 12V down to those voltages pushed onto them.

We may see the deletion of 5V from the ATX spec once SATA finally dies, but that is still far down the road. We would need an entirely new PCIe-type spec that wholly omits 3.3V for that voltage to die, and I'm not seeing that happen anytime soon.