AMD Sempron (Kabini) Quad Core AM1 @ $37

[coercitiv]

Not sure. But don't think its a passive vs. active cooling thing.

I don't think it's a passive vs active cooling either.

Static leakage should not be an issue for BT at idle, that's what power gating is there to do.

 

[sm625]

What is the price of that 15W Haswell ??? Those Haswell SKUs are not in the same segment as Kabini.

You can find an ivy bridge ULV pentium in a tablet for under $400 at amazon By summertime those deals should be replaced by the haswell version.

 

[Enigmoid]

Or a chip not as good as on other sites , remember that all chips do not use the same voltages so some chips can consume somewhat more in quite
a large variability, that said i prefer numbers from a relevant set up than
from THG who seems to produce hasardous not to say rigged results,
just look at the recent power numbers of BT in the AM1 review and
in a previous review using a BT that was supposed to consume less :

They did their job very well , i hope that they were rewarded
accordingly for such efforts.

Criticism lands where criticism is due.

Here is the correct graph. You have total power consumed not the power usage of the platform at one instant in time.

 

[Abwx]

Criticism lands where criticism is due.

Here is the correct graph. You have total power consumed not the power usage of the platform at one instant in time.

Granted but technicaly speaking i find their set ups irrelevant, i much prefer hardware.fr who measure both the main and the 12V rail and publish both values wich allow to see how innefficient either MB or PSUs can be while isolating the CPU/APU + board VRMs losses total power , other methodologies as Anand using only the main power with 1200W PSUs is a laugh for whom has the slightest technical or engeenering training, the dutch site was the closest on the posted list of a professional review by the virtue of their 150W PSU , still oversized but not to the point that its eventual losses at iddle are 100% if not more than the tested plateform supposed iddle comsumption and 50% or so when it is loaded...

 

[Enigmoid]

Granted but technicaly speaking i find their set ups irrelevant, i much prefer hardware.fr who measure both the main and the 12V rail and publish both values wich allow to see how innefficient either MB or PSUs can be while isolating the CPU/APU + board VRMs losses total power , other methodologies as Anand using only the main power with 1200W PSUs is a laugh for whom has the slightest technical or engeenering training, the dutch site was the closest on the posted list of a professional review by the virtue of their 150W PSU , still oversized but not to the point that its eventual losses at iddle are 100% if not more than the tested plateform supposed iddle comsumption and 50% or so when it is loaded...

I agree. Each site has its own strengths and weaknesses.

 

[Abwx]

I agree. Each site has its own strengths and weaknesses.

I forgot that Anand didnt publish power numbers for thev AM1 review
but it s his usual way of proceding, we ll see in the next part of the
review where we ll get thoses numbers.

While we re at it here THG s 850W PSU efficency in function of power output, at 10% , that is 85W and well above the power consumed by the plateforms, efficency is awfull and even more nose diving at the levels measured by THG.

http://www.tweakpc.de/hardware/tests/netzteile/xfx_pro850_core_edition/s03.php


Edit : On another note , Kabini support ECC RAM.

http://developer.amd.com/wordpress/media/2012/10/48751_BKDG_Fam_16h_Mod_00h-0Fh.pdf

 

[positivedoppler]

I wonder how overclockable kabini is

 

[daveybrat]

I wonder how overclockable kabini is

Pretty sure they aren't overclockable unfortunately.

 

[VirtualLarry]

http://www.newegg.com/Product/ComboD...=Combo.1596228

112$. I'd say 27$ would be worth it here.

I cannot fathom why either this AMD platform or the Bay Trail BGA platform cost nearly 90$. What a ripoff for what you get, considering the alternative that doesn't cost much more.

That combo IS a good deal. But what if the user wants to use XP? Haswell rigs won't run XP, Intel hasn't released (and likely won't) drivers for XP.

AMD was at least claiming in their slides that the AM1 platform supports XP, although I read one report that XP drivers for those APUs weren't available either, but that was pre-launch, so that might have been why.

 

[blackened23]

That combo IS a good deal. But what if the user wants to use XP? Haswell rigs won't run XP, Intel hasn't released (and likely won't) drivers for XP.

AMD was at least claiming in their slides that the AM1 platform supports XP, although I read one report that XP drivers for those APUs weren't available either, but that was pre-launch, so that might have been why.

Hah. What. Maybe the better question is what kind of self loathing PC user would want to use Windows XP.

Windows 8 versus Windows 7. Now THAT I can see. There are good and valid reasons for holding out on Windows 8, although the upcoming service pack for Windows 8.1 sounds very promising. (I say this as someone who likes 8.1) But Windows XP? Win XP belongs in a museum along with rotary phones and betamax VCRs. Now I obviously used XP ages ago and it was good for its time. Just like Windows 3.11 was good for its time. At this point though, it's pretty unbelievable. It's been nearly 15 years.

To each their own, though. My main point was that the price on this AMD platform is not nearly attractive enough to the budget crowd to make it appealing over the Haswell celeron and pentium CPUs. You can easily get those at 90-100$, and they do perform significantly better than both this AMD platform and Bay Trail. While costing about the same.

I really think something needs to give in terms of pricing on both of those latter platforms compared to Haswell Celeron / Pentium. 10-20$ doesn't warrant the absolutely huge performance difference between the two platforms.

 

[VirtualLarry]

Hah. What. Maybe the better question is what kind of self loathing PC user would want to use Windows XP.

I was mainly speaking about those that would have a business use case to continue to use the XP platform.

To each their own, though. Clearly some differences in thought among everyone on this.

Sure, for us enthusiasts, XP should have been dead and gone by the time Win7 was released. At least, it was for me. But some people actually have a reason to continue to use XP.

 

[shady28]

The 5350+AM1 is $93, and thats what im talking here, the Sempron Quad+AM1 mb is about $74, Sempron dual will be at 50 or something?

VS 5350

-J1900 is about the same on CPU, uses less power and its pasive, about the same price too, but inferior in gaming.
-G1820 is better all around, but uses more power, it also cost about $10 more.



Only the pclab review is good.

First of all, you guys need to price ***Mini ITX*** builds.

Then you need to check out how well a Haswell CPU cooler fits in a Mini-ITX case.

I find it amusing how this discussion went from comparing Kabini to Bay Trail, then after the 5350 review everyone wants to compare Kabini to Haswell.

To be blunt, that tells me that Kabini is a great chip.

 

[SPBHM]

First of all, you guys need to price ***Mini ITX*** builds.

Then you need to check out how well a Haswell CPU cooler fits in a Mini-ITX case.

I find it amusing how this discussion went from comparing Kabini to Bay Trail, then after the 5350 review everyone wants to compare Kabini to Haswell.

To be blunt, that tells me that Kabini is a great chip.

well, on my case I always compared both with Haswell or the 1037U, and I think for most users Haswell would be better, better performance, better platform with more expansion and upgrade, power usage on this level is not a big deal for an OK case.
now between BT and Kabini I would go with BT for a tablet, Kabini for a laptop or desktop for sure, but with the option for Haswell Celeron, the choice would be easy...

AM1 will also be used for regular sized cases on cheap OEM PCs and so on, so I think it's very valid to compare it to haswell celeron.

now if you can add a significant difference to haswell+h81 price without going with a dual core Kabini, it becomes more interesting, but I'm afraid 1.3GHz and no OC is to low performance, still should be OK for a NAS, but only 2 sata ports for a NAS?!

 

[VirtualLarry]

Wondering how the 5350 stacks up to a B75 / G1610 combo, for daily-driver usage? Web, online video, no gaming or rendering.

Also wondering how it stacks up to an Athlon II X4 630 / 640 (2.8, 3.0Ghz)?

I used to pick up those combos at MC for $100 with an Athlon II X4 quad-core, and FREE motherboard. I thought that was one of the greatest deals since sliced bread, at the time.

Currently, MC is offering the FX-6300 with free mobo for like $110 + tax in-store.

Just to keep all of this Kabini hype in perspective, as far as price / performance goes.

Edit: Also as a replacement for an Athlon64 X2 4800+ S939 with 2GB of RAM, and some lower-end older discrete video card. Web browsing, no gaming.

 

[pw257008]

And they've got tray FX-4130s for $39.99, though no combo on it, so with cheap mobo it comes to around the price of 5350+mobo. (of course, without IGP)

 

[monstercameron]

comparing deep discounts on -near- eol products with price of a newly released product...might aswell be be buying stuff off ebay.

 

[pw257008]

comparing deep discounts on -near- eol products with price of a newly released product...might aswell be be buying stuff off ebay.

just pointing out nice deals at microcenter

 

[cbn]

comparing deep discounts on -near- eol products with price of a newly released product...might aswell be be buying stuff off ebay.

A lot of it is still being sold new though at many retailers.

For example, just this week I bought a new ECS BGA Celeron 847 thin-mini itx board at Newegg for only $32 After rebate, free shipping.

Same goes with LGA 1155. I bought an ECS Z77 ATX with 32GB mSATA SSD for $50 after rebate too, With my G1620 it comes out to less than $95 for the three item combo.

So in many ways AMD's and Intel's new tech must still be competitive agaist the old tech,

Either the prices drops for the same performance or the performance (and features) need to increase if the price stays the same.

 

[raghu78]

Undoubtly but it s still 28nm and the claimed numbers for Beema/Mullins would be quite possible with this same node if it was only for the benchs scores that were published , thing is that not only perfs are better but TDP is dramaticaly reduced , that s too much for a tweak that would use the same process.

Temash was built on TSMC 28HP while Mullins is likely built at TSMC 28HPM. So right there is an opportunity to reduce active power due to a more power efficient process

http://www.tsmc.com/english/dedicatedFoundry/technology/28nm.htm

"TSMC also provides high performance for mobile applications (HPM) technology to address the need for applications requiring high speed as well as low leakage power. Such technology can provide better speed than 28HP and similar leakage power as 28LP. With such wide performance/leakage coverage, 28HPM is also ideal for many applications from networking, tablet, to mobile consumer products. "

Temash lacked support for deeper sleep states and its idle power was too high.

http://seekingalpha.com/article/2134173-amd-the-new-puma-chip-as-a-catalyst-in-mobile

This article gives an explanation as to how Mullins and Beema achieve their stated numbers. anandtech clearly stated that AMD was conservative with the TDP ratings for Temash/ Kabini. With Mullins/Beema AMD is being more accurate with the TDP ratings.