• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Haswell to require new Power Supply?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
no anti intel.

just the nastly truth no body want to hear or acknowledge. afterall haswell (or ethanol) is suppose to be green!

Wait, this is not true. They're reducing the wattage for longer battery life. The feature is easily disabled if necessary and doesn't need to impact you at all. Additionally, pretty much every new big-brand PSU supports it (even older ones will as long as they weren't crappy $20 ones).
 
The original story at vr-zone is completely wrong.

Exactly. It was a retarded article by vr-zone to trigger some clicks and drama. And the usual crowd jumped right in without any form of knowledge or understanding of the issue.

Exactly how is it completely wrong and retarded?

I just read about this over on JonnyGuru. Found a nice post in the thread too. check it

Post 3 seems to be funny.. read the last line

I'm not sure he understands what's going on, or I don't understand what he is saying. The Intel slide clearly states that there is a new C7 state that is entered even if the display is active, which I interpret as 'while using your pc', not as 'while pc is in sleep state'. Or maybe he does mean 'while using your pc, but when you do something that requires full cpu power it will take as long as a reboot before you can'.

Anyway, I think I'll skip Haswell but if I didn't I would want to use its power saving features to the fullest.
 
Wait, this is not true. They're reducing the wattage for longer battery life. The feature is easily disabled if necessary and doesn't need to impact you at all. Additionally, pretty much every new big-brand PSU supports it (even older ones will as long as they weren't crappy $20 ones).

Exactly, and that is what numerous other articles and follow-ups are now saying since the initial rumor came out. If you have a new(1-2 years depending upon manufacturer), quality PSU you should be fine.
 
It is hard to believe that for any modern power supply, the minimum load on a 12V rail has to be 0.5A per ATX v2.3 specs. That is 6 watts! No way. Even 0.05A is still more than half a watt.
 
Most of existent PSU are incompatible with Haswell. Some Haswell-certified PSUs start to appear.

Yes, BIOS disabling of C6/C7 could be a solution, but this will increase power consumption.
 
Most of existent PSU are incompatible with Haswell. Some Haswell-certified PSUs start to appear.

Yes, BIOS disabling of C6/C7 could be a solution, but this will increase power consumption.

I think you misunderstand what the word incompatible means. Existing PSU are compatible with Haswell.
 
I think you misunderstand what the word incompatible means. Existing PSU are compatible with Haswell.

I think you failed to obtain my point. Fortunately others got it and remark how most PSUs in the market are incompatible with Haswell

http://www.bit-tech.net/news/hardware/2013/05/01/haswell-psu/1

http://www.xbitlabs.com/news/cpu/di...y_Require_New_Power_Supply_Units_for_PCs.html

http://www.tomshardware.com/news/Intel-Haswell-PSU-Power-State,22320.html

http://news.softpedia.com/news/Here...mpatible-with-Intel-Haswell-CPUs-349994.shtml
 
Last edited:
I didn't know anything about the power states, found this which helped me: http://impact.asu.edu/cse591sp11/Nahelempm.pdf

A little old but I couldn't find anything else with as broad coverage of the different power states and with simple explanations.

Some of the articles about Haswell and the new power states are misleading, at least in title, which seems to be why everybody is flipping out. I guess it depends on how you interpret "(in)compatible with Haswell". The titles like "Haswell's new C6/C7 power states may have issues with current PSU's" certainly reads a lot different than "Haswell may require a new PSU". Its the difference between missing out on a power-saving feature and potentially having to buy a new power supply.

I would like to see a comparison of power usage between Ivy Bridge and Haswell in each of the major power states. Mostly just want to see how the lowest idle states on Haswell compare to OS sleep on Ivy Bridge, and if you disable those states, do you end up using more or less power with Haswell. I suppose it would be nice if you could just walk away, never have to bother putting the OS to sleep, and still use as little or nearly as little power.
 
The fact that Enermax even list legacy (2008) PSU products as Haswell compatible should start to make a few people think.

But again, it would remove the drama and make this thread useless.
 
Last edited:
A little old but I couldn't find anything else with as broad coverage of the different power states and with simple explanations.

http://www.hardwaresecrets.com/arti...out-the-CPU-C-States-Power-Saving-Modes/611/1

The fact that Enermax even list legacy (2008) PSU products as Haswell compatible should start to make a few people think.

But again, it would remove the drama and make this thread useless.

Seasonic doesn't list S12 and M12 psu's as compatible, there are still a lot of those in use.
 
Last edited:
Fortunately others got it and remark how most PSUs in the market are incompatible with Haswell
That statement clearly qualifies as FUD. There is currently zero evidence to support your "incompatible" claim. First off, this is still a rumor. Second, if the rumor turns out to be true, disabling the C6 and C7 power states will fix the issue.

Most of existent PSU are incompatible with Haswell. Some Haswell-certified PSUs start to appear.

Yes, BIOS disabling of C6/C7 could be a solution, but this will increase power consumption.
For someone who appears to be an avid AMD enthusiast, why would a tiny increase in power consumption concern you on a desktop computer? I've seen you downplay AMD's power consumption problems in the past so this argument seems ironic to me.
 
This reminds me of all the different memory that 'definitely doesn't work' if it's not listed on the MB maker's website as 'compatible'.

This whole story reeks of FUD...
 
That statement clearly qualifies as FUD. There is currently zero evidence to support your "incompatible" claim. First off, this is still a rumor. Second, if the rumor turns out to be true, disabling the C6 and C7 power states will fix the issue.

First, there are four links (bit-tech, xbitlabs, tomshardware, and softpedia) in my post all them saying the same that I am saying. You snipped all them in your response.

Second, you don't need to repeat what I already said about BIOS and C6/C7.
 
A PSU not being compliant with the new sleep states is not the same thing as being incompatible with Haswell.
 
First, there are four links (bit-tech, xbitlabs, tomshardware, and softpedia) in my post all them saying the same that I am saying. You snipped all them in your response.

Second, you don't need to repeat what I already said about BIOS and C6/C7.

The 4 links you posted are reiterating the same rumor. That does not constitute evidence for most power supplies being incompatible with Haswell.

But hey, you want to diss on Haswell. Fine.
 

"no clues if their own power supply will be compatible with the new energy functions of Intel Haswell CPUs."

Pretending that this means the power supplies are not compatible with the CPU's is akin to saying USB3.0 devices are not compatible with USB2.0 ports. They are compatible, you just won't be able to take advantage of some of the new features.
 
dule-popcorn.gif
 
"no clues if their own power supply will be compatible with the new energy functions of Intel Haswell CPUs."

That is from the first of the four links given by me. Now don't stop and continue reading... and you will find "and thus are unlikely to be compatible with Haswell".

Evidently the four links were written for general public, and not for some hyper-sensible Intel enthusiasts...
 
Look, I don't think that there are actually many PSUs that can use Haswell's new power states. But that is a very big difference from saying that PSUs are not "compatible" with Haswell.
 
That is from the first of the four links given by me. Now don't stop and continue reading... and you will find "and thus are unlikely to be compatible with Haswell".

Evidently the four links were written for general public, and not for some hyper-sensible Intel enthusiasts...

No, it was simply poorly written. Most of us recognize it, you don't. It has nothing to do with brand enthusiasm, it has to do with accurate/factual information, which neither the articles, nor your presentation of the articles are. The publishers have an excuse, that being, sensationalistic headlines generate more hits, more hits generate revenue. Whats your excuse? You were politely corrected, but insisted on continuing upon the path of ignorance and are now getting defensive about it.
 
Last edited:
That is from the first of the four links given by me. Now don't stop and continue reading... and you will find "and thus are unlikely to be compatible with Haswell".

Evidently the four links were written for general public, and not for some hyper-sensible Intel enthusiasts...

You don't need to be an Intel enthusiast to understand where you've gone wrong. All you have to do is have access to a dictionary.
 
Look, I don't think that there are actually many PSUs that can use Haswell's new power states. But that is a very big difference from saying that PSUs are not "compatible" with Haswell.

Yeah, that's like saying AMD's AM3+ motherboards don't support the Southern Island cards because they don't have PCIe 3.0.

Actually, it does work 100% of the time. You just lose out on a tiny bit of performance, maybe.
 
Back
Top