No, not really. I think the point was that comparing CPUs whilst running GPU constrained settings and, moreover, doing so only through a "very scientific" method based on attendants to a party playing for a bit, is a very natural thing to do when you have poor CPUs. Because that scenario dampens the differences in CPUs.
itsmydamnation
Veteran
meh, I think AMD need to do more of this stuff, because the reality is the advantages intel have over AMD matter less then the advanatges AMD have over intel to the average punter. AMD need to change the topic of the conversation.
Last edited by a moderator:
Let's take it in a different direction, then...
If you were to poll the users at that gaming expo to identify how many of them use a dedicated video card versus use an IGP, how many would claim the IGP?
Without question, the IGP in the AMD is far superior to Intel's current offerings. Does that make the AMD IGP a viable option for the gaming enthusiasts at that expo? Perhaps some of them will be happy to play at resolutions, quality settings and a framerate that are amicable to the AMD IGP.
I wouldn't. However, I am likely not in AMD's target demographic for this platform.
AMD's newest line of processors will be the most interesting to people wanting a cost-effective yet reasonably capable laptop gaming system. In fact, we had a few testing units floating around the office last year (search for my prior posts, you'll find me talking about it) and I really liked the performance. If I were in the market today for a new laptop, a Llano 3850 would be right up my alley -- maybe even with the hybrid crossfire.
For laptops or desktops that are aimed at light casual gaming (flash,HTML5, that sort of thing) Intel and AMD options are going to deliver a similar experience. There simply isnt' enough performance demand at the 'light' level to need all that IGP awesomeness.
Now, in terms of more advanced gaming, I'm not convinced that a huge number of people are going to opt to keep / use the AMD IGP and forego the dedicated card. In my personal interpretation, most of the 'moderate' gamers will pop for a low cost midrange dedicated card, which will wipe the floor with the AMD IGP.
That's my own spin on the situation
If they were blind they wouldn't be buying APUs with good graphics now, would they?
I wonder how AMD expects to get 10% higher performance from Piledriver. According to that Software Optimization Guide, there's really not that many performance enhancing changes. Ok 10% more entries in load/store queue (44 instead of 40) is nice, as is the doubled l1 dtlb size (from 32 to 64). There's also a couple more supported instructions but otherwise it seems virtually unchanged - even keeping the very lame l1i associativity of Bulldozer for both Trinity and Vishera.
Of course you can't directly estimate performance by this guide but since everything seems to be so extremely similar I really don't expect much (except maybe higher clocks).
Of course you can't directly estimate performance by this guide but since everything seems to be so extremely similar I really don't expect much (except maybe higher clocks).
That's the thing I'm not sure there are really "performance bugs". Sure there are performance issues but it's unclear if these are really the result of bugs as such and not just unbalanced architecture. (I would have considered the low li1 associativity a likely candidate for causing performance issues but apparently amd didn't think so...)
Maybe some changes like being able to execute some more instructions in the AGUs go in the right direction, but those are not in Piledriver.
Maybe some changes like being able to execute some more instructions in the AGUs go in the right direction, but those are not in Piledriver.
I think someone said the leaked die shot showed extra SRAM in the area of the branch predictor. I'm not entirely sure.
The bulk of the improvement seems to be around reducing the power draw of the cores and better clocks. Llano had some notable weaknesses here that Piledriver's more advanced turbo and circuit implementations should improve upon.
Llano's inflexible turbo option doesn't set the bar that high.
The bulk of the improvement seems to be around reducing the power draw of the cores and better clocks. Llano had some notable weaknesses here that Piledriver's more advanced turbo and circuit implementations should improve upon.
Llano's inflexible turbo option doesn't set the bar that high.
Yes, that area is where the branch prediction logic resides. Probably the BTB size has been increased, along with other still unknown tweaks.
Last edited by a moderator:
I thought Trinity was 32nm.
According to the SOG all family 15h processors have the same number of BTB entries.
That's why I said there doesn't really seem to be many architectural changes, the SOG really has quite some details. But yes, better clocks / turbo could help quite a bit.
So it'll be on par with Phenom 2 and around 3 generations behind Intel latest and Greatest.
At this pace Intel will be a good 60%+ faster per clock.
Well they are saying a 17watt Trinity will be as fast as a 30w Llano. That should put it pretty close to Sandy bridge IMO.
The jury is still out on what increases if any we will see with Ivery bridge
Leaked trinity over at overclock.net show that its no faster then Llano