More Llano leaks (A8 APU extensively benchmarked)

[nonameo]

Techreport has their review out. I want to see reviews of production systems. Where are they?

http://techreport.com/articles.x/21099

 

[Idontcare]

Really? How do we know that exactly?

I thought this was public domain info?

If its not, then, uhh, yeah what I meant to say was "speculated as being"...:hmm:

All these subliminal messages really bring into question your impartiality.

Ah, again with the personal attacks I see. Can't deftly attack the message so you got to go below the belt and do the whole attack the messenger spiel?

 

[busydude]

Really? How do we know that exactly? All these subliminal messages really bring into question your impartiality.

No.. it brings you AMD fanboyism into light.

Techreport has their review out. I want to see reviews of production systems. Where are they?

http://techreport.com/articles.x/21099

2nd week of July IIRC.

 

[AtenRa]

http://www.youtube.com/watch?v=XcENbpO7mhw&feature=player_embedded

 

[piesquared]

No.. it brings you AMD fanboyism into light.

I didn't think that was ever in question. I'm also a fanboy of the Vancouver Canucks (shame on the fans), Ferrari, Chevy and many other things in life.

As for hitting below the belt, I call them as I see them and don't get fooled easily anymore. Some things are too obvious.

As for techreport's article, why don't more reviews investigate the abysmal image quality like Scott did? Anandtech's article makes sandy britche's 'graphics' look wayyyyy better than it acually is. Suppose framerates were compared at equal image quality, sandy bridge would be even more laughable than it is today. Of course it's not possible to increase intel's image quality, but could at least reduce AMD's image quality to get it at least closer to the same level. AMD and NV have had to play by those rules for 10+ years, now intel wanders in and get's a pat on the bum even though it's graphics are abysmal? And people wonder why so many have a disdain for anything intel. They don't play by the same rules, they change them to fit their agenda. Vista is just one of the many many examples.

Radeon



HD3000






Image quality was important until intel enters the picture. Pathetic.

 

[Idontcare]

As for hitting below the belt, I call them as I see them and don't get fooled easily anymore. Some things are too obvious.

If it's obvious then it doesn't need to be said.

And regardless how you feel about it, stating it as you did is a violation of the posting rules.

There is no "the ends justifies the means" get out of jail free clause in the guidelines.

 

[Idontcare]

As for techreport's article, why don't more reviews investigate the abysmal image quality like Scott did? Anandtech's article makes sandy britche's 'graphics' look wayyyyy better than it acually is. Suppose framerates were compared at equal image quality, sandy bridge would be even more laughable than it is today. Of course it's not possible to increase intel's image quality, but could at least reduce AMD's image quality to get it at least closer to the same level. AMD and NV have had to play by those rules for 10+ years, now intel wanders in and get's a pat on the bum even though it's graphics are abysmal? And people wonder why so many have a disdain for anything intel. They don't play by the same rules, they change them to fit their agenda. Vista is just one of the many many examples.

Radeon



HD3000






Image quality was important until intel enters the picture. Pathetic.

^ that's why I'm replacing this Intel IGP laptop with a Llano laptop as soon as possible

With Intel IGP not only are you rewarded with dodgy graphics drivers that bomb out your games every now and then, but during those lucky timeperiods where you do get to play your slideshow of a game the IQ is crap.

 

[busydude]

As for techreport's article, why don't more reviews investigate the abysmal image quality like Scott did?

I agree with you one IQ part.. and frankly I was surprised that Anand did not mention IQ.

He mentioned the quality was not up to mark in Asymmetric crossfire configurations, but never talks about IQ comparisons. There is an obvious difference in IQ between Intel's IGP and Fusion.. fairly disappointed I have to say.. regarding Anand and Jarred's review.

 

[Mopetar]

Has anyone conducted a study on how many people attempt to game on just the Intel HD graphics. I'm assuming that the people who own those notebooks are mostly just playing Flash games in the browser. I would think that most people who are interested in playing more demanding titles are purchasing notebooks with discreet graphics.

Llano (and IB depending on performance) may change that to some extent, but the number of people trying to gaming on a SB notebook without a discreet card has to be fairly low. Either that or they're mostly playing older titles like the original Fallout or Civilization II. Is anyone aware of a study examining this in further detail?

 

[Vesku]

People make do with what they can afford, it's one of the reasons flash and 2D games do surprisingly well. Swap out peoples' terrible IGP for decent graphics and I bet you'd see usage change pretty quickly. Also software companies would feel less constrained on the graphics side of development.

 

[AnandThenMan]

I'm starting to wonder if this isn't going to be another 90nm->65nm type of transition where the 65nm chips could hardly clock as high as their 90nm older siblings.

I would sure like to know how you even began to think this way. GloFo's 32nm process is excellent, and if you look at what the 3500M is offering for the power budget, performance is fantastic.

We already know Bulldozer was officially delayed because of lackluster clockspeed yields

Citation needed.

I agree with you one IQ part.. and frankly I was surprised that Anand did not mention IQ.

IQ was not tested for some of the same reasons higher graphics settings and resolutions were not tested, DX11 mode was not tested etc.

 

[Idontcare]

I would sure like to know how you even began to think this way. GloFo's 32nm process is excellent, and if you look at what the 3500M is offering for the power budget, performance is fantastic.

If you got some performance/watt review data to throw my way and help educate me so my opinion becomes aligned with yours then by all means, hook me up.

Citation needed.

http://www.brightsideofnews.com/news/2011/5/30/amd-delays-bulldozer-citing-performance-issues.aspx

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

 

[AnandThenMan]

If you got some performance/watt review data to throw my way and help educate me so my opinion becomes aligned with yours then by all means, hook me up.

http://techreport.com/articles.x/21099

Take a look at the battery life, and also the overall performance. The actual user experience is the best so far for a mobile device IMO.

http://www.brightsideofnews.com/news/2011/5/30/amd-delays-bulldozer-citing-performance-issues.aspx

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

I don't see any official announcements regarding the reasons for the delay (if there is one) anywhere from AMD in those links, maybe I missed them?

 

[Idontcare]

I don't see any official announcements regarding the reasons for the delay (if there is one) anywhere from AMD in those links, maybe I missed them?

See my response above to piesquared, I had my timeline confused regarding public/nonpublic infos.

I meant to say "speculated" not "official".

And thanks for the link :thumbsup: Tech report is in my usual rotation of review sites, I missed this one. Reading it now.

 

[Accord99]

On the other hand, the previewed desktop Llano's power consumption is unimpressive, being higher than the 2500K:

http://www.anandtech.com/show/4447/desktop-llano-motherboards-the-asrock-a75-extreme6-preview/3

 

[Phynaz]

GloFo's 32nm process is excellent

GloFo's 32nm is so good they just fired something like three top executives, including their CEO.

 

[richierich1212]

On the other hand, the previewed desktop Llano's power consumption is unimpressive, being higher than the 2500K:

http://www.anandtech.com/show/4447/desktop-llano-motherboards-the-asrock-a75-extreme6-preview/3

That's because Anand tested it with a 5850. Do you think most people buying Llano for desktop systems would also buy a discrete video card as well?

 

[Cerb]

Has anyone conducted a study on how many people attempt to game on just the Intel HD graphics. I'm assuming that the people who own those notebooks are mostly just playing Flash games in the browser. I would think that most people who are interested in playing more demanding titles are purchasing notebooks with discreet graphics.

Llano (and IB depending on performance) may change that to some extent, but the number of people trying to gaming on a SB notebook without a discreet card has to be fairly low. Either that or they're mostly playing older titles like the original Fallout or Civilization II. Is anyone aware of a study examining this in further detail?

http://techreport.com/articles.x/21099/13

Not to be confused with high detail and 100W+ cards, nor i5/i7 w/ discrete mobile GPUs, but the playability is there for many games (I wouldn't call 20-40 FPS enough for shooters, but the rest are fine).

 

[Accord99]

That's because Anand tested it with a 5850. Do you think most people buying Llano for desktop systems would also buy a discrete video card as well?

There were also power measurements using the integrated graphics and regardless of the video card, the similarly equipped 2500K system used less power under load.

 

[CTho9305]

^ that's why I'm replacing this Intel IGP laptop with a Llano laptop as soon as possible

With Intel IGP not only are you rewarded with dodgy graphics drivers that bomb out your games every now and then, but during those lucky timeperiods where you do get to play your slideshow of a game the IQ is crap.

Do you actually notice the image quality outside of directed tests?

 

[richierich1212]

There were also power measurements using the integrated graphics and regardless of the video card, the similarly equipped 2500K system used less power under load.

Ok but I don't see the point of you bringing this up. Since the Llano has a faster iGPU we should assume it would consume more power considering the 2500K is rated for 95w, while the Llano is rated for 100w.

 

[Idontcare]

Do you actually notice the image quality outside of directed tests?

I (attempt to) play the same games on my laptop that I play on my desktop...save for DAO which I won't even bother to install on the laptop.

So I am looking at the game through two different LCD's and two different GPU's. Resolution is different too but not crazy different. (19x12 vs 14x9)

The drop in IQ when playing on my laptop is just pitiful. Now I can handle the pitiful stuff, its the buggy graphic artifacting quirks that get to me.

Sure you learn to live with it and to work around it, but the laptop is well past the stage of having one foot in the grave and I'm not going to intentionally make the same mistake twice.

And I'm also aware that today's Intel IGP is likely no where near as sucky as the aging one I aim to replace, but since I've got a choice and I already threw good money after what turned out to be a shoddy product I'm just not interested in giving it another chance this time around.

 

[Accord99]

Ok but I don't see the point of you bringing this up. Since the Llano has a faster iGPU we should assume it would consume more power considering the 2500K is rated for 95w, while the Llano is rated for 100w.

Well, the IGP is supposed to gated off when using discrete video cards. The main point is that Llano is not proof that GloFlo's 32nm process is excellent as when clock speeds are pushed higher, it so far appears to not have better absolute power consumption, let alone performance/watt then Intel's 32nm products.

 

[msroadkill612]

I think AMD would prefer a 25watt dual core too (& one is planned), but the 4 core was easier to bring to market. Switching nodes is tricky enough.

Brazos was all new also, but on a familiar node.

A 28nm shrink for brazos would be a killer product.

 

[msroadkill612]

Actually When 28nm is released I am going to buy NV gpu . I bought a 580 . First NV gpu in 6,7 years for me. Alot of SB buyers will use NV . Have you seen the differance in performance On highend gpus on SB comparred to all other cpus . SB kicks butt. BY alot comparred to using with a stars like CPU. Your assuming Xfire will work great on APU cross fired. Just a couple of days left than we can talk facts.

I saw some CF results somewhere & they were terrible. Pity but not surprising. Early drivers etc. & may be using crap ram

may have been the same review that showed intels display quality problems.