• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Launches Three Kaveri APU SKUs in February 2014 – Feature Set For A10 and A8 APUs Detailed

AMD A10-7850K Performance Review.

Alongside the A10-7700K, the A10-7850K is a recent addition to AMD's A-Series line of APUs and uses the existing FM2+ socket. By using the new 28nm manufacturing process, these new APUs are able to run faster than the previous generation, even when running at a lower clock frequency. In addition, improvements in the graphical portion of the APU increases the capabilities of the integrated R7 graphics controller. Interestingly, the clock speed of these new APUs are actually lower than the models they are replacing.

http://www.pugetsystems.com/labs/articles/AMD-A10-7850K-Performance-Review-529/

ssEi5JG.jpg
NpGgMj9.jpg
 
Last edited:
The A10-7850K is 10% slower than the A10-6800K. or am i going daft?

It's FPU performance which AMD hasn't been great at, other results show gains even with the lower clock, FPU's the one for gaming performance though.

But Kaveri's lower clocked remember... Seems the improvements depends on the task (And people blast Intels consistent IPC improvements for being meager :p)

I'll hold judgement, looking like it's not going to be consistent though, so it can go from having nigh on Lynnfield IPC to lower.

Wait for reviews and see how it fairs.
 
It's FPU performance which AMD hasn't been great at, other results show gains even with the lower clock, FPU's the one for gaming performance though.

But Kaveri's lower clocked remember... Seems the improvements depends on the task (And people blast Intels consistent IPC improvements for being meager :p)

I'll hold judgement, looking like it's not going to be consistent though, so it can go from having nigh on Lynnfield IPC to lower.

Wait for reviews and see how it fairs.

I think i'm going to wait for release reviews from the main reviewers, so far we have had 3 or 4 unknown reviewers using the same Cinebench R15 and all are showing completely different results.

Seems they are all fighting for attention by pulling results out of their backsides for a chip they don't even have.
 
Last edited:
My only hesitation is that the results won't really be focusing on the CPU cores performance, rather its GPU (Or posting results that aren't of a consistency, like Trinitys preview with the cores touting improvements which were a far cry off reality)

Either way, I think we'll have to wait for end user reviews.
 
The A10 7850K looks an overpriced pile of fail. At that price you can get an Athlon II X4 760K and an HD7750 or HD7770. Only the A8 7600 withs it configurable 45W/65W TDP looks interesting.

Someone at AMD must be smoking skunk to think they can charge £130 to £140 for an A10 7850K. Supposedly there is noise BF4 will be included with the A10 7850K and even then it makes it only slightly better value than an A10 6800K. The problem is that A10 6800K looks like a faster CPU at stock and I suspect it can be overclocked far more too. So we are left with the IGP only being the real improvement with the A10 7850K and even that looks rubbish.

The only think that Kaveri can do better is GPU compute. That is the ONLY major improvement it seems.

The jump from Llano to Trinity on the same node was greater regarding gaming performance,especially with the IGP.

If this stays true in other tests then Kaveri deserves to be flamed by reviewers and even the fanbois out there. I hope at least power consumption is closer to the A10 6700 otherwise this should have been an OEM only release.

It seems their marketeers have not learned from the Bulldozer fiasco. If this review is truely indicative of performance I can see the A10 7850K price crashing hard.

AMD should have just stuck with 384 shaders and devoted more of the TDP to raising the clockspeeds of the CPU section.

FFS,even make it a 3 module,6 thread part??

This was what Kaveri was meant to be according to earlier leaks of official documents.

Desktop Kaveri looks stillborn. FFS,the AMD marketeers are foolish. They should have shown reviews of the mobile and lower TDP parts first. An example is the A8 7600 in 45W TDP mode which has 600MHZ higher base clockspeed than the A10 6700T.

However,by showing the higher TDP desktop parts first,it makes it look like a sidegrade.

FFS,even Intel IB review samples had smaller dies than retail samples,which probably made it look better regarding power consumption.

I really hope AMD Carrizo is out by the end of the year. If Intel plonks down GT3 as their standard IGP in the next CPU refresh,AMD are in big trouble on the desktop.

This is looking like a worse launch than Bulldozer.

At least Bulldozer was new microarchitechure,with room to grow and the performance delta between the Phenom II X6 and FX CPUs has improved as software support has got better.

However,this won't be the case of Steamroller over Piledriver I suspect.
 
Last edited:
@ CAT, if Broomsticks latest review is true then yes its an utter fail, a total pile of junk.

If this was even priced at £100 to £110,you could call it an improved A10 6800K,but its going to cost £120+ it seems.

Those CB scores are the same as A10 5800K running at 3.8GHZ ot 4.2GHZ,and the A10 7850K runs at 3.7GHZ to 4GHZ!!

What is the bet it still has the same fail cooler meaning the A10 7850K does not Turbo and the A10 6800K does??

Doesn't AMD even bother to run benchmarks?? FFS,tweak the flipping CPU in some way,so that it beats its predecessor in single thread tasks FFS,or even matches it.

Its like they don't care and yet they will send this out to multiple reviewers on will beat on about it.

Do they even read the reviews??

They are being paid to this job and how can they not fathom how it will look??

What is wrong with AMD marketeers and the people involved in product positioning?? Forget Intel,existing products from AMD themselves are making this look poor value for money and a sidegrade at best.
 
AMD are historically bad at pricing products that look to be anywhere good.

I touched on the pricing earlier :p

Lets wait until the 14th before we bash it :p

But then we had Bulldozer and the leaks and the "Don't believe it" from AMD personnel themselves, and that was a bag of crap.
 
Bit early for ranting about supposed benchmarks. There's a few anomalous results in there too R.E. RAM speed/timings that need looking at in a proper review.

The problem is that Puget Systems is a system integrator like OcUK but in the US and reasonably well known for decent systems. The sample they tested came from AMD themselves,and AMD allowed them AFAIK to publish the benchmarks.

The CB R15 single threaded and multi-threaded scores more or less matches the previous leak we saw with another sample from a far eastern source.

You are talking about 90 points for the single threaded score and around 300 to 310 points for the multi-threaded score.

The scores are broadly similar to an A10 5800K and Athlon II X4 760K,which are around £85 and £60 ATM.

The IGP scores are not that great and remember the other SKUs have 384 shaders instead of 512 too,so unless their are some driver issues(quite possible),it might mean the Trinity and Richland IGPs are faster than the 384 shader versions,which at best cost the same,but are probably going to cost more.

In that review the IGP performance improvement is tiny. To put it in context,Trinity was a much larger improvement in IGP performance and it was on the same process node as Llano.

The only real improvement is in compute.

I really,really hope the reviews next week,show a much bigger performance improvement over Richland,especially as the A10 7850K looks like it will cost around £130 to £140. It really needs that free copy of BF4 which is supposed to be bundled with it if the leaks are true.

If the pricing were the same it would not be such an issue,but sadly it isn't.
 
Last edited:
Back
Top Bottom