AMD A10-5800K preview - iGPU side only

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

NostaSeronx

Diamond Member
Sep 18, 2011
3,811
1,290
136
Okay, yeah memory is the bottleneck not the ROPs. ROPs performance is directly linked to the Radeon Memory Bus. So, Llano and Trinity are memory bus bottlenecked.

Hopefully, those Kaveri rumors with 256 MB GDDR5 caches across 32-bit ports are true.
 

Eeqmcsq

Senior member
Jan 6, 2009
407
1
0
I thought these previews were to only show GPU performance?
How Xbit get away with SYSmark2012 bench?

http://www.xbitlabs.com/articles/graphics/display/amd-trinity-graphics_5.html

sysmark.png

From the techreport article where they opted NOT to do a preview:

http://techreport.com/blog/23638/amd-attempts-to-shape-review-content-with-staged-release-of-info

Non game benchmarks
- Traditional benchmarks are designed to highlight differences in different architectures and how they perform. We understand that this is a useful tool for you and that your readers expect to see this data. The importance of these results is in your evaluation, as the leading experts, of what these performance numbers mean. We encourage you to use your analysis if you choose to publish a preview article and if you find that to be appropriate to your approach to that article. The numbers themselves must be held until the October [redacted] embargo lift. This is in an effort to allow consumers to fully comprehend your analysis without prejudging based on graphs which do not necessarily represent the experiential difference and to help ensure you have sufficient content for the creation of a launch day article.

They got away with it by not posting actual CPU numbers.
 

inf64

Diamond Member
Mar 11, 2011
3,884
4,692
136

SPBHM

Diamond Member
Sep 12, 2012
5,067
422
126
Hmm interesting find. Maybe they posted it by mistake? But 5800K outperforming 4170 is a great improvement. 4170 has ~6% higher clock so effectively PD is ~12% faster than BD in sysmark . Note that sysmark is BS benchmark though.

this result and the power consumption (of the CPU running p95, not the entire APU) are looking quite promising compared to llano and bulldozer,

Okay, yeah memory is the bottleneck not the ROPs. ROPs performance is directly linked to the Radeon Memory Bus. So, Llano and Trinity are memory bus bottlenecked.

Hopefully, those Kaveri rumors with 256 MB GDDR5 caches across 32-bit ports are true.

DDR4 may help a lot also...
the cache thing is interesting, AMD had something like it on their AM2/AM3 IGPs, but it was to slow (32bit low clocked DDR2/DDR3), it worked... but the increase was not amazing (but... it was a 40sps GPU lol)

there's the rumors that intel will adopt a 128mb cache for their fastest haswell IGP 1066MHz but with a very wide bus, going for 60+ Gbps!
 

notty22

Diamond Member
Jan 1, 2010
3,375
0
0
The problem with the desktop chips/igp's being tested at laptop resolutions is the non-native screen resolution on your desktop monitor is going to look, 'blah'. And at medium settings,probably magnified.
 

ShintaiDK

Lifer
Apr 22, 2012
20,378
146
106
Llano is memory limited too.

For any iGPU development to kick off we need ondie GPU memory or 256bit DIMMs. I really hope they move away from 64bit DIMMs with DDR4. S2011 reminds me of the old Pentiums and Ppro with 4x EDO RAM. And thats not positive.
 

taltamir

Lifer
Mar 21, 2004
13,576
6
76
AMD has allowed some desktop Trinity benchmarks to be released a week early, but only on theiGPU

Oh so THATS why the reviewers were only benchmarking the iGPU and ignoring the CPU part entirely. And here I was coming to ask why anandtech's latest review was so one dimensional.
 

ShintaiDK

Lifer
Apr 22, 2012
20,378
146
106
http://techreport.com/blog/23638/amd-attempts-to-shape-review-content-with-staged-release-of-info

Ouch.

We are allowing limited previews of the embargoed information to generate additional traffic for your site, and give you an opportunity to put additional emphasis on topics of interest to your readers. If you wish to post a preview article as a teaser for your main review, you may do so on September 27th, 2012 at 12:01AM EDT.
The topics which you are free to discuss in your preview articles starting September 27th, 2012 at 12:01AM EDT are any combination of:
- Gaming benchmarks (A10, A8)
- Speeds, feeds, cores, SIMDs and branding
- Experiential testing of applications vs Intel (A10 Virgo will be priced in the range of the i3 2120 or i3 3220)
- Power testing
We believe there are an infinite number of interesting angles available for these preview articles within this framework.
We are also aware that your readers expect performance numbers in your articles. In order to allow you to have something for the preview, while maintaining enough content for your review, we are allowing the inclusion of gaming benchmarks.
By allowing the publication of speeds, feeds, cores, SIMDs and branding during the preview period, you have the opportunity to discuss the innovations that AMD is making with AMD A-Series APUs and how these are relevant to today’s compute environment and workloads.
In previewing x86 applications, without providing hard numbers until October [redacted], we are hoping that you will be able to convey what is most important to the end-user which is what the experience of using the system is like. As one of the foremost evaluators of technology, you are in a unique position to draw educated comparisons and conclusions based on real-world experience with the platform.

The topics which you must be held for the October [redacted] embargo lift are:
- Overclocking
- Pricing
- Non game benchmarks

Non game benchmarks
- Traditional benchmarks are designed to highlight differences in different architectures and how they perform. We understand that this is a useful tool for you and that your readers expect to see this data. The importance of these results is in your evaluation, as the leading experts, of what these performance numbers mean. We encourage you to use your analysis if you choose to publish a preview article and if you find that to be appropriate to your approach to that article. The numbers themselves must be held until the October [redacted] embargo lift. This is in an effort to allow consumers to fully comprehend your analysis without prejudging based on graphs which do not necessarily represent the experiential difference and to help ensure you have sufficient content for the creation of a launch day article.

We are aware that this is a unique approach to product launches. We are always looking at ways that we can work with you to help drive additional traffic to your articles and effectively convey the AMD message. We strive to provide the best products in their price points, bringing a great product for a great price. Please feel free to provide feedback on what you find, both with the product and with your experience in the AMD New Product Review Program. We try to ensure that we are providing you what you need and appreciate any feedback you have to offer on how we can do better.

I think we all know what that translate into for the CPU part. And possible overclocking too.
 
Last edited:

inf64

Diamond Member
Mar 11, 2011
3,884
4,692
136
We know the CPU performance already(THG review posted months ago),so what is your point? We know the OC margin for CPU cores also,it's 4.5-4.6Ghz on air cooling and maybe more on some higher end coolers. There is no mystery.
 
Aug 11, 2008
10,451
642
126
The Tech Report refuses to take part: http://techreport.com/blog/23638/amd-attempts-to-shape-review-content-with-staged-release-of-info

Must be a stinker.

I don't really understand Trinity's target market. If I'm not gaming, Intel graphics are good enough. If I'm really gaming, Trinity graphics don't cut it. Does a significant middle market actually exist (outside of laptops where this makes more sense)?

I agree. The GT640 ddr3 is pretty much reviled every time it is mentioned. Somehow trinity, which loses to it in basically every test, is now this great gaming chip. I dont get it.

I could not figure it out from the rambling post from the AMD representative, but did AMD specify which cards Anand could test against? If so, I am surprised he would even buy into that. Since the HD7750 is the fastest card that doesnt require an external power connecter, and could be put into almost any off the shelf system, that is the card I would have liked to see Trinity compared to. I am sure AMD did not want that data shown, since the HD7750 would make the igpu look even worse.

Trinity, or any APU, might have a niche market in all in ones and HTPCs, but it is not a chip I would ever consider for gaming on the desktop. Period.
 

Olikan

Platinum Member
Sep 23, 2011
2,023
275
126
I agree. The GT640 ddr3 is pretty much reviled every time it is mentioned. Somehow trinity, which loses to it in basically every test, is now this great gaming chip. I dont get it.

well, the 640 cost $100 buck alone...good luck finding a better cpu for $30
 

beginner99

Diamond Member
Jun 2, 2009
5,318
1,763
136
I don't really understand Trinity's target market. If I'm not gaming, Intel graphics are good enough. If I'm really gaming, Trinity graphics don't cut it. Does a significant middle market actually exist (outside of laptops where this makes more sense)?

Agree. The middle market thing would make sense if getting a significantly better setup would be very expensive but it's not. The contrary is the case.

Also the advantage of the GT 640 seems to increase with higher resolution. Who has a 720p desktop monitor in 2012?
 

inf64

Diamond Member
Mar 11, 2011
3,884
4,692
136
well, the 640 cost $100 buck alone...good luck finding a better cpu for $30
I think many of them just forget the fact that they need to actually pay for the add-on(discrete) GPU part and that it takes additional slot ,produces heat/consumes energy. In their reality it costs nothing,consumes next to nothing in terms of energy .Oh it also produces no sound and takes no space in the PC case. It's practically free :p .
 

Mallibu

Senior member
Jun 20, 2011
243
0
0
Agree. The middle market thing would make sense if getting a significantly better setup would be very expensive but it's not. The contrary is the case.

Also the advantage of the GT 640 seems to increase with higher resolution. Who has a 720p desktop monitor in 2012?

I agree, even though its nice for an iGPU to be making progress it's still useless to most since it's in a "grey" area of performance which is "jack of all trades" but still very low for anything particular. It's overkill for anyone who doesn't play games and useless for anyone who plays since they're miles better with a cheap discrete GPU. Performance needs to be 2x-3x higher to start entering entry gamers setups.
 
Mar 6, 2012
104
0
0
Agree. The middle market thing would make sense if getting a significantly better setup would be very expensive but it's not. The contrary is the case.

Also the advantage of the GT 640 seems to increase with higher resolution. Who has a 720p desktop monitor in 2012?

You could always just set the resolution to 720p if a game don't run in 1080p well enough. If you're buying an apu you're probably not a graphics whore anyways so if the gameplay is good enough that should matter the most.
A significantly better setup may not be very expensive, but it is more expensive none the less. I'd much rather avoid the "hassle" of a discrete gpu, both out of form factor and because it shouldn't be necessary. But trinity isn't quite strong enough yet for my liking, so I won't be getting it. Maybe haswell or kavari will let me get a tiny pc which I can use for everything.
 
Aug 11, 2008
10,451
642
126
well, the 640 cost $100 buck alone...good luck finding a better cpu for $30

Well, right now, i5 systems are 100 to 200 dollars cheaper in big box stores than an A10. The particular ones I saw were in costco. They had an i5 levano, I believe ivy bridge even, for 599.00, while an HP A10 was 799.00.

So, put in the 100 dollar HD7750 and you are still 100 dollars cheaper than the A10, and have a computer that is much faster in both CPU and GPU performance and will be more future proof. You could even pay somebody 50.00 to put in the GPU and be ahead.

But for the sake of agrument, I will accept your contention that the discrete intel system is 100 dollars more expensive. If you keep the system 3 years, that is less than 3.00 per month. You could skip a game or two and save that much up front, or just skip one McDonalds meal per month. Heck, you would not even have to skip the games. Just wait till they are on a steam sale and save the money.
 
Aug 11, 2008
10,451
642
126
I agree, even though its nice for an iGPU to be making progress it's still useless to most since it's in a "grey" area of performance which is "jack of all trades" but still very low for anything particular. It's overkill for anyone who doesn't play games and useless for anyone who plays since they're miles better with a cheap discrete GPU. Performance needs to be 2x-3x higher to start entering entry gamers setups.

Well said. As a matter of fact, I have used the exact same description in earlier posts regarding APUs. "Jack of all trades, and master of none" actually is the full saying, and it describes APUs very well, especially in a desktop.

APUs might have a place in laptops if you only want to play some older, less demanding games occasionally, but not in the desktop.
 

ultimatebob

Lifer
Jul 1, 2001
25,134
2,450
126
Heh... the CPU performance must be pretty lousy if they're trying to get all of the tech review sites to post articles about GPU performance first.

Maybe they're hoping that people will put their pre-orders in before the CPU benchmarks come out.

If I was Anand, I would have held back my review instead of releasing a partial review that is potentially misleading.
 

sm625

Diamond Member
May 6, 2011
8,172
137
106
From the Legit Reviews article:

The ambient air temperature in the room was 75F or 24C. The Gigabyte EasyTune6 software showed that our CPU had a temperature of 17C at idle with the fan spinning at ~1750 RPM. This is odd as it was below the room temperature, but remember CPU voltage is derived from the voltages inside the processor, so things are always a bit funky with temperatures.

Seriously? This is the first time I've seen anyone even attempt to explain this, and it sounds awfully hokey.
 

Hacp

Lifer
Jun 8, 2005
13,923
2
81
Well, right now, i5 systems are 100 to 200 dollars cheaper in big box stores than an A10. The particular ones I saw were in costco. They had an i5 levano, I believe ivy bridge even, for 599.00, while an HP A10 was 799.00.

So, put in the 100 dollar HD7750 and you are still 100 dollars cheaper than the A10, and have a computer that is much faster in both CPU and GPU performance and will be more future proof. You could even pay somebody 50.00 to put in the GPU and be ahead.

But for the sake of agrument, I will accept your contention that the discrete intel system is 100 dollars more expensive. If you keep the system 3 years, that is less than 3.00 per month. You could skip a game or two and save that much up front, or just skip one McDonalds meal per month. Heck, you would not even have to skip the games. Just wait till they are on a steam sale and save the money.

Or you could skip a McDonalds meal per month, wait for a steam sale, buy the A10, and still be 100 dollars ahead.
 

Hacp

Lifer
Jun 8, 2005
13,923
2
81
Heh... the CPU performance must be pretty lousy if they're trying to get all of the tech review sites to post articles about GPU performance first.

Maybe they're hoping that people will put their pre-orders in before the CPU benchmarks come out.

If I was Anand, I would have held back my review instead of releasing a partial review that is potentially misleading.

We already know the CPU numbers. 12-15% better than Bulldozer. Plus, you can overclock your chip to 4.4 on air and blow the I3s out of the water. Ya, and you don't need to spend 120 bucks on an expensive motherboard either.