A8-3850 Has Ineffective BClk Multiplier

[btarunr]

"Empty Overclocking" is a term we just made up, to describe unreal overclocking headroom that does not translate into any performance improvements, with AMD's A8-3850 APU. This chip can be set to run at base clock multiplier value above 29x on some motherboards, that will increase clock speed being reported to you, but that "increased" clock speed will not translate to any performance improvements at all.

This means that the multiplier is ineffective in driving the clock above its maximum default value. So the next time you see screenshots screaming something like "6.00 GHz" on air with the base clock at its default 100 MHz, don't be fooled, trust only those overclocking feats in which the multiplier is set at the maximum default (29.0x) or lesser, and in which the overclocker has increased the base clock among other things.



Update: It seems like AMD is aware of the issue, and forewarned reviewers about it. Apparently a glitch in the BIOS code allows the users to "set" higher multiplier values than the chip can respond to, even as the chip doesn't run at those values. Utilities like CPU-Z read those BIOS-set values and display the effective clock speed, even as the actual clock speed doesn't budge. AMD recommends only the base clock increase method for overclocking. As always, AMD warned that overclocked chips are not covered by product warranties. Perhaps future BIOS updates by motherboard vendors will fix this bug.

Source: OCWorkbench, Image Courtesy Newegg.com

[RejZoR]

Ok, is it just me or is this a bit erm, dumb? Why don't they just hard lock it if there is no point in increasing it anyway?

[jmcslob]

Negative attention is still Attention.

[Chevalr1c]

Well, AFAIK it is not a BE/FX/what shall I call it. So people won't buy it b/c of the unlocked MP, I guess. So the few fools who will "OC" this CPU via the MP will be easy to spot and ID as idiots (one should not trust OC reports without BClk/MP details anyway).

[arterius2]

well, what is the maximum default?

[PHaS3]

"to describe tests confirm to be unreal overclocking headroom that does not translate into any performance improvements at all"

What does not translate is that sentence... Maximum English confusion... is maximum

[btarunr]

Originally Posted by arterius2

well, what is the maximum default?

29.0x

[Maban]

Llano/Lynx isn't even supposed to have an unlocked multiplier as far as I'm aware.

[Enmity]

Originally Posted by Chevalr1c

Well, AFAIK it is not a BE/FX/what shall I call it. So people won't buy it b/c of the unlocked MP, I guess. So the few fools who will "OC" this CPU via the MP will be easy to spot and ID as idiots (one should not trust OC reports without BClk/MP details anyway).

This, most enthusiasts should know what to look for if they want to overclock - in the case of Bulldozer this will be the black edition/ FX models. Much like intels K series being monster overclockers and non k's suckin worse than a Maori hooker.

[xaira]

i read somewhere that this dued was saying that all bulldozer ES suffer from the same thing so any clock for clock comparsons with sandy at 4.0ghz where sandy is like crushing BD by like 25% that thats good news for amd because all current overclocks on new models is empty until the bug is fixed so that 4ghz sandy is actually going against the stock BD

[Trackr]

So I could take it up to 500x and it would show up as 50.0Ghz?

That's worth the money just to see.

[btarunr]

Originally Posted by Trackr

So I could take it up to 500x and it would show up as 50.0Ghz?

That's worth the money just to see.

That's about as useful as using photoshop to make a 50 GHz screenshot.

[btarunr]

Added an update.

[erixx]

Kudos for bringing more 'conflictive' news to the frontpage, and not only post P.R. letters of many times boring products like usb sticks!!!

BIG CHEERS!

[HalfAHertz]

Um pretty sure this was known since review day. I think Xbit labs were the first to report the glitch and they got a confirmation from Asrock as well...Not really news worthy.

[Haytch]

I find this all rather interesting.

I personally wouldn't get one, but that doesn't mean it's going to fail.
99% of my clients dont O.C.

If a BIOS level update resolves this glitch, then . . . It's all good.

[spynoodle]

Oh, AMD..... when will you finally catch up to Intel?

[63jax]

PcPer just made an article about overclocking on Llano A8 and IT does make a difference, up to 30 percent performance win.

[Dent1]

Originally Posted by spynoodle

Oh, AMD..... when will you finally catch up to Intel?

Catch up in what respect? Explain yourself!

[JATownes]

Originally Posted by 63jax

PcPer just made an article about overclocking on Llano A8 and IT does make a difference, up to 30 percent performance win.

Source Link?

[caleb]

PentiumII behaved the same way. Even if you OC'd with jumpers above the multi it was still locked at its stock. Apparently its locked.

[nINJAkECIL]

Originally Posted by spynoodle

Oh, AMD..... when will you finally catch up to Intel?

I thought AMD had already past Intel.
oh wait.....
that's GPU. Sorry.

[DigitalUK]

just a bios glitch, the last bios "glitch" amd had gave many users extra cores for free.

[63jax]

Originally Posted by JATownes

Source Link?

http://www.pcper.com/news/Graphics-C...ng-Performance

[TheLaughingMan]

I thought this was NOT a BE chip and there where no addition multipliers above the 29x? I remember reading somewhere quote, "...with the multiplier, you can go down, not up."