The HUGE question. Please ask AMD about that. DO CARRIZO COME WITH COLOR COMPRESSION LIKE TONGA? And if yes, that's for all of them, all the models?
Having a Sempron 3850 I am disappointed to see that AMD abandoned that socket. While the SOC is great for an HTPC and a few other things thanks to those four cores, it does have it's limitations like HEVC playback. A Beema would have been a nice upgrade.
Another thing to point here is that those two core parts, AMD should throw them in the dust bin and BURN THEM. It is utterly stupid to market those processors for use in PCs or laptops. You do make a few dollars from those SOCs, financially it looks correct, but the end result is someone getting a brand new, last generation dual core processor that it is going to disappoint him big time. So AMD's image is really destroyed at that person's eyes. Someone who buys an Intel Atom will blame himself for buying a low performing Intel cpu. Intel are the fastest right? So he will fix that by buing a faster Intel CPU. Someone who buys a dual core low end AMD, will blame AMD for the performance and will never again choose AMD. Will lock itself at Intel. People see the same problem from a different perspective based on the brands recognition and the general opinion the public has about a company's products.
I'm not gonna be back here, so listen well. I shall say this only once. You are part of the problem with public perception by nagging AMD for making low cost parts. As you describe in the end (i assume your view is the same). When someone buys a weak AMD CPU it's AMD's fault, but when they buy from Intel it's his/her own fault. How does that make sense? If you bought low end AMD CPU try a better one next time, it will cost less then an i5 from Intel. That's what i did, and it still runs without problems.
Not being back here again means that you would not realize how much you misunderstood what I posted. This is a nice way to throw BS all over the place and never realize it. A stupid and arrogant approach of the internet.
I am really happy from my unlocked 445, my main 1055 and the 3850 in the HTPC, thank you very much.
When you buy the cheap firm's product and you are disappointed, you blame the cheap firm and you never buy again from the cheap firm. When you buy from the premium brand and you are not satisfied, there is a big chance you will buy again from the premium brand, this time a more expensive product, because most of the others will tell you that the premium brand in this case, cpu performance, is the only option you have.
It seems to me that he has a perfect understanding of what you said. Essentially what you said was:
If I you buy from AMD and you are disappointed then you blame AMD and buy from Intel.
If you buy from Intel and you are disappointed the you blame yourself and buy from Intel again.
Personally, If I buy from AMD and I am disappointed, I buy a faster, more powerful chip from AMD. If I buy from Intel and I am disappointed, I buy from AMD. I then pocket the huge savings.
Intel braggs a lot as to how much better they are than AMD. I find that Intel is very expensive compared to AMD. So, if I am dissatisfied with intel, I just buy a faster, more powerful AMD processor and put the difference in my pocket.
I have been very happy with the AMD APU's and find that the Intel comparable products are lackluster and grossly more expensive.
It allows them to compete in the business market for design wins, whereas before they were off the table. I am curious to see the implementation. At my job we encrypt all hard drives and leverage the TPM. Nothing enters our doors without a TPM that is running MS Windows.
The cpu design team for the cat cores is supposedly gone from AMD, anyone know the back-story? Jaguar has been a huge success in the semi-custom (consoles, ...) business.
All I know is roughly during the time Jim Keller returned to AMD, John Bruno (the chief architect of APU in the PC/notebook world inside AMD), left for Apple...
But in the corporate world, when strategic decisions/future roadmaps are made, it is a difficult place to be in as funding would be cut for the deprecated teams. And ppl from there would be either laid off or re-assigned thus becoming a shell of what once was a big team.
I have been in the receiving end where my platform was not the "selected" choice going forward to be replaced by another "new platform". The funding was drastically cut.
We beat the odds and did a lot of deliverables, POCs with a lot of success and proved that we can meet the new platform goals admirably and beat it even. Also the new platform did not bring in the ROI and was found to be "Lost the forest for the trees" concept.
Only we proved what we could do we started to get continuous funding but still I have to justify any new hires....
I don't think they are gone. Cat won't be continued. Which was foreseeable, IMHO. Because everything will be Zen from 2016 on. The engineers probably switched to others design teams, for example Zen/K12.
From the leaks found especially at Geekbench database (e.g. http://browser.primatelabs.com/geekbench3/2113126 ), suggests that Carrizo-L is most likely rebranded Mullins/Beema APUs (had the same family, model and stepping) with slightly higher frequency. That 'Radeon' is most likely R2, like those found on E2-6110 and E1-6010.
From http://www.anandtech.com/show/7314/intel-baytrail-... mentions "The power consumption, at least on the CPU side, also looks very good. From our SoC measurements it looks like Bay Trail’s power consumption under heavy CPU load ranges from 1W - 2.5W, putting it on par with other mobile SoCs that we’ve done power measurements on.". Thus that is definitely around 2W as mentioned earlier. AMD's Carrizo-L has much higher wattage of around 10W to 25W thus AMD's Beema TDP range (although Mullins and Beema uses the same chip). There are no AMD Mullins APUs with 2W SDP. The lowest is 2.8W SDP as shown here http://www.anandtech.com/show/7974/amd-beema-mulli... for both the quad core and dual core versions. Despite that "2.8W SDP", so far all the tablets using AMD's Mullins APU are at least 10 inches and bigger (including the AMD Discovery http://www.anandtech.com/show/7974/amd-beema-mulli... at 11.6 inch, and Bungbungame Photon 2 http://www.bungbungame.com/EN/products/products.as... at 10.1 inch).
2 GCN CUs seems really anemic, even if it was for a smartphone SoC. 10W puts it into large tablet territory, but again with 2CUs, they're either clocked at >1GHz or the thing may not even compete with Cherry Trail in GPU power.
If neither Kabini in 2013 and Beema in 2014 couldn't get any traction. why bother launching what it seem to be the same chip with a couple of non-substantial tweaks? So what's the point? Is this just to keep investors "happy" claiming they're launchign "new chips" every year?
2 GCN CUs is WAY more graphics power than Atom has. Atom's have 4 EU iirc. 8 GCN units beats 20 GCN... 2 then should beat 5... So, this is 25%+ faster than the 4 EU Atoms.
BayTrail Atoms had 4 EUs. And this would be way more than 25% faster than those Atoms. However, CherryTrail Atoms have 16 EUs (albeit at a lower clock), which might pose quite a challenge (in practice the CherryTrail gpu seems to be easily twice as fast as the one from BayTrail). I suspect they could be close in the end, though probably only because they won't quite compete in the same TDP class (BayTrail atoms even in desktop format didn't go past 10W, whereas this is 12-25W).
Cherry Trail also has the advantage of dual channel memory. It could be that Carrizo-L gets it as well (due to sharing the socket with Carrizo), but considering AMDs recent track record of (almost) not changing anything I wouldn't get my hopes up.
Ah, though they stayed at 4! Oops. Thanks. Yeah, Cherry should be faster, but, at the power it would be faster, well... There are better options for iGPU from either vendor at that point.
But...but...these aren't in Atom's class! Even if the performance is similar or even a little better, these are 10w on the lowest end and 25w on the high end. Atom is a 2w SDP processor, maybe clocking 4w TDP at most. Even the notebook/desktop chips are only about 11w TDP and that is being generous. You are talking double to quadruple the TDP.
The 10w dual core Puma+ might be honest, but at best the top end quad core Intel Atom chip is maybe hitting the 11w TDP, the dual core Celeron Atom parts are almost certainly hitting a more realistic 6-7w at most.
These are not targetting Atom at all. They are maybe targetting Broadwell ULT, but I suspect not really on performance.
To further my thought, last I checked, the Puma design is not significantly better IPC than Silvermont cores...and the dual core Celeron desktop/notebook designs are hitting 40+% higher clock speeds than the Puma+ dual core part...at a realistic 30-40% less power consumption.
Hard to judge AMD's products until we can get power consumption numbers. At least, for the cat cores. Brazos at "18W" ran rings around a "9W" Atoms and used less power.
Of course, that was the old atom and the bobcat core, but, the point is, need to see some systems with these in them first!
I expect Intel to have lower power, higher pref/watt and AMD have higher absolute performance. What that will lead to in realistic scenarios where you are racing to sleep depends on the near-idle power usage.
Problem: the cat cores never had good and agressive turbo. Which is why they either have "large" TDPs or low clock speeds, but nothing like Silvermonts turbo giving +50% single threaded performance.
So, can someone tell me in simple words what this all means for me? Specifically, I'm ignoring anything Atom and even Core M, Y etc due to pathetic GPUs. I do not want a step back from the Intel HD 4400 of my Surface Pro 2. Now that Broadwell is out, I guess the new reasonable minimum, or the thing to beat for AMD is the HD 5500. Can AMD offer better GPU performance in a similar power envelope? And will someone finally make a decent tablet with their part?
Sorry, but no, 15W TDP is not too high. I even mentioned a tablet that has similar CPUs in it - Surface Pro 2. Pro 3 also has variants with i5 with a HD 4400, despite also having weaker i3 variants.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
50 Comments
Back to Article
nandnandnand - Tuesday, May 12, 2015 - link
"28nm"triggered
nandnandnand - Tuesday, May 12, 2015 - link
Bulldozer's last gasp. Pin your hopes on Zen instead.Novacius - Tuesday, May 12, 2015 - link
What. There's no Bulldozer in Carrizo-L.testbug00 - Tuesday, May 12, 2015 - link
This isn't Bulldozer based...patrickjp93 - Tuesday, May 12, 2015 - link
Excavator is still part of the bulldozer or construction family depending on which circles you speak to.PixyMisa - Tuesday, May 12, 2015 - link
Carrizo-L is a Bobcat family chip, while Carrizo is Bulldozer family.movax2 - Tuesday, May 12, 2015 - link
Correct. Bulldozer has 1 module with two cores.While Bobcat family has just small cores, not modules.
http://en.wikipedia.org/wiki/List_of_AMD_CPU_micro...
testbug00 - Tuesday, May 12, 2015 - link
this is not Bulldozer, Piledriver, Steamroller, or Excavator. Or any other Construction Equipment core from AMD.gonchuki - Monday, May 18, 2015 - link
Don't be so sure about that: http://www.bobcat.com/excavators/Bulldozer/Bobcat originated as a word play for representing the big vs small cores.
yannigr2 - Tuesday, May 12, 2015 - link
The HUGE question. Please ask AMD about that.DO CARRIZO COME WITH COLOR COMPRESSION LIKE TONGA? And if yes, that's for all of them, all the models?
Having a Sempron 3850 I am disappointed to see that AMD abandoned that socket. While the SOC is great for an HTPC and a few other things thanks to those four cores, it does have it's limitations like HEVC playback. A Beema would have been a nice upgrade.
Another thing to point here is that those two core parts, AMD should throw them in the dust bin and BURN THEM. It is utterly stupid to market those processors for use in PCs or laptops. You do make a few dollars from those SOCs, financially it looks correct, but the end result is someone getting a brand new, last generation dual core processor that it is going to disappoint him big time. So AMD's image is really destroyed at that person's eyes. Someone who buys an Intel Atom will blame himself for buying a low performing Intel cpu. Intel are the fastest right? So he will fix that by buing a faster Intel CPU. Someone who buys a dual core low end AMD, will blame AMD for the performance and will never again choose AMD. Will lock itself at Intel.
People see the same problem from a different perspective based on the brands recognition and the general opinion the public has about a company's products.
Novacius - Tuesday, May 12, 2015 - link
Carrizo (not Carrizo-L) will very likely have Delta Color Compression. It's GCN 1.2/v3.SleepyFE - Tuesday, May 12, 2015 - link
I'm not gonna be back here, so listen well. I shall say this only once. You are part of the problem with public perception by nagging AMD for making low cost parts. As you describe in the end (i assume your view is the same). When someone buys a weak AMD CPU it's AMD's fault, but when they buy from Intel it's his/her own fault. How does that make sense? If you bought low end AMD CPU try a better one next time, it will cost less then an i5 from Intel. That's what i did, and it still runs without problems.yannigr2 - Tuesday, May 12, 2015 - link
Not being back here again means that you would not realize how much you misunderstood what I posted. This is a nice way to throw BS all over the place and never realize it. A stupid and arrogant approach of the internet.I am really happy from my unlocked 445, my main 1055 and the 3850 in the HTPC, thank you very much.
When you buy the cheap firm's product and you are disappointed, you blame the cheap firm and you never buy again from the cheap firm. When you buy from the premium brand and you are not satisfied, there is a big chance you will buy again from the premium brand, this time a more expensive product, because most of the others will tell you that the premium brand in this case, cpu performance, is the only option you have.
JHS28677 - Tuesday, May 19, 2015 - link
It seems to me that he has a perfect understanding of what you said. Essentially what you said was:If I you buy from AMD and you are disappointed then you blame AMD and buy from Intel.
If you buy from Intel and you are disappointed the you blame yourself and buy from Intel again.
Personally, If I buy from AMD and I am disappointed, I buy a faster, more powerful chip from AMD. If I buy from Intel and I am disappointed, I buy from AMD. I then pocket the huge savings.
Intel braggs a lot as to how much better they are than AMD. I find that Intel is very expensive compared to AMD. So, if I am dissatisfied with intel, I just buy a faster, more powerful AMD processor and put the difference in my pocket.
I have been very happy with the AMD APU's and find that the Intel comparable products are lackluster and grossly more expensive.
lefty2 - Tuesday, May 12, 2015 - link
Does anyone actually use the AMD Secure Processor feature? Just curious.eanazag - Tuesday, May 12, 2015 - link
It allows them to compete in the business market for design wins, whereas before they were off the table. I am curious to see the implementation. At my job we encrypt all hard drives and leverage the TPM. Nothing enters our doors without a TPM that is running MS Windows.AMD has other hurdles now at 28nm designs.
R3MF - Tuesday, May 12, 2015 - link
any hope of seeing these in socket AM1? they are listed as up to 25W, rather than carefully tiered in TDP as the bga products were...Novacius - Tuesday, May 12, 2015 - link
They seem to be configurable by the vendor (12-25W).Spoelie - Tuesday, May 12, 2015 - link
David Kanter made an interesting comment at the tech report:http://techreport.com/discussion/28246/the-tr-podc...
The cpu design team for the cat cores is supposedly gone from AMD, anyone know the back-story? Jaguar has been a huge success in the semi-custom (consoles, ...) business.
testbug00 - Tuesday, May 12, 2015 - link
Er, i watched that whole thing and never heard that part 0.o? Any time stamps or close to when he discussed it?silverblue - Tuesday, May 12, 2015 - link
He mentioned it in the comments section.MrSpadge - Tuesday, May 12, 2015 - link
This might explain why there are no new cat-cores on the roadmaps shown at the recent FAD.testbug00 - Wednesday, May 13, 2015 - link
Ah. Thanks.rocketbuddha - Tuesday, May 12, 2015 - link
All I know is roughly during the time Jim Keller returned to AMD, John Bruno (the chief architect of APU in the PC/notebook world inside AMD), left for Apple...But in the corporate world, when strategic decisions/future roadmaps are made, it is a difficult place to be in as funding would be cut for the deprecated teams. And ppl from there would be either laid off or re-assigned thus becoming a shell of what once was a big team.
I have been in the receiving end where my platform was not the "selected" choice going forward to be replaced by another "new platform". The funding was drastically cut.
We beat the odds and did a lot of deliverables, POCs with a lot of success and proved that we can meet the new platform goals admirably and beat it even. Also the new platform did not bring in the ROI and was found to be "Lost the forest for the trees" concept.
Only we proved what we could do we started to get continuous funding but still I have to justify any new hires....
gruffi - Wednesday, May 13, 2015 - link
I don't think they are gone. Cat won't be continued. Which was foreseeable, IMHO. Because everything will be Zen from 2016 on. The engineers probably switched to others design teams, for example Zen/K12.krumme - Tuesday, May 12, 2015 - link
Contrarevenue vs Oil - Fight !BlueBlazer - Tuesday, May 12, 2015 - link
From the leaks found especially at Geekbench database (e.g. http://browser.primatelabs.com/geekbench3/2113126 ), suggests that Carrizo-L is most likely rebranded Mullins/Beema APUs (had the same family, model and stepping) with slightly higher frequency. That 'Radeon' is most likely R2, like those found on E2-6110 and E1-6010.movax2 - Tuesday, May 12, 2015 - link
Well, that's not bad.I think new Mullins are interesting too. Their performance was on par with fastest Atom Z3xxx family.
It should very good for tablets.
BlueBlazer - Wednesday, May 13, 2015 - link
Except chips for fanless tablets are typically around 2W (like Intel's Bay Trail http://www.anandtech.com/show/7314/intel-baytrail-... ), while Carrizo-L is 10W to 25W range. Large fanless tablets around 10 inch and larger can use 4.5W chips (like Intel's Broadwell-Y http://www.anandtech.com/show/9160/the-dell-venue-... ) but that's is still way below Carrizo-L's wattage. Furthermore its not faster than Bay Trail-T actually when compared to a quad core tablet SoC like Intel Atom Z3736F http://browser.primatelabs.com/geekbench3/compare/... or compared to a dual core 4.3W SoC like Intel Atom N2807 http://browser.primatelabs.com/geekbench3/compare/... Thus Carrizo-L will most probably will be destined for low cost laptops rather than tablets...gruffi - Wednesday, May 13, 2015 - link
2W is SDP, not TDP. There are also Mullins APUs with 2W SDP. Don't mix SDP with TDP! That's completely misleading.BlueBlazer - Wednesday, May 13, 2015 - link
From http://www.anandtech.com/show/7314/intel-baytrail-... mentions "The power consumption, at least on the CPU side, also looks very good. From our SoC measurements it looks like Bay Trail’s power consumption under heavy CPU load ranges from 1W - 2.5W, putting it on par with other mobile SoCs that we’ve done power measurements on.". Thus that is definitely around 2W as mentioned earlier. AMD's Carrizo-L has much higher wattage of around 10W to 25W thus AMD's Beema TDP range (although Mullins and Beema uses the same chip). There are no AMD Mullins APUs with 2W SDP. The lowest is 2.8W SDP as shown here http://www.anandtech.com/show/7974/amd-beema-mulli... for both the quad core and dual core versions. Despite that "2.8W SDP", so far all the tablets using AMD's Mullins APU are at least 10 inches and bigger (including the AMD Discovery http://www.anandtech.com/show/7974/amd-beema-mulli... at 11.6 inch, and Bungbungame Photon 2 http://www.bungbungame.com/EN/products/products.as... at 10.1 inch).ToTTenTranz - Tuesday, May 12, 2015 - link
2 GCN CUs seems really anemic, even if it was for a smartphone SoC.10W puts it into large tablet territory, but again with 2CUs, they're either clocked at >1GHz or the thing may not even compete with Cherry Trail in GPU power.
If neither Kabini in 2013 and Beema in 2014 couldn't get any traction. why bother launching what it seem to be the same chip with a couple of non-substantial tweaks?
So what's the point? Is this just to keep investors "happy" claiming they're launchign "new chips" every year?
testbug00 - Tuesday, May 12, 2015 - link
2 GCN CUs is WAY more graphics power than Atom has. Atom's have 4 EU iirc. 8 GCN units beats 20 GCN... 2 then should beat 5... So, this is 25%+ faster than the 4 EU Atoms.mczak - Tuesday, May 12, 2015 - link
BayTrail Atoms had 4 EUs. And this would be way more than 25% faster than those Atoms. However, CherryTrail Atoms have 16 EUs (albeit at a lower clock), which might pose quite a challenge (in practice the CherryTrail gpu seems to be easily twice as fast as the one from BayTrail). I suspect they could be close in the end, though probably only because they won't quite compete in the same TDP class (BayTrail atoms even in desktop format didn't go past 10W, whereas this is 12-25W).MrSpadge - Tuesday, May 12, 2015 - link
Cherry Trail also has the advantage of dual channel memory. It could be that Carrizo-L gets it as well (due to sharing the socket with Carrizo), but considering AMDs recent track record of (almost) not changing anything I wouldn't get my hopes up.testbug00 - Tuesday, May 12, 2015 - link
Ah, though they stayed at 4! Oops. Thanks.Yeah, Cherry should be faster, but, at the power it would be faster, well... There are better options for iGPU from either vendor at that point.
azazel1024 - Tuesday, May 12, 2015 - link
But...but...these aren't in Atom's class! Even if the performance is similar or even a little better, these are 10w on the lowest end and 25w on the high end. Atom is a 2w SDP processor, maybe clocking 4w TDP at most. Even the notebook/desktop chips are only about 11w TDP and that is being generous. You are talking double to quadruple the TDP.The 10w dual core Puma+ might be honest, but at best the top end quad core Intel Atom chip is maybe hitting the 11w TDP, the dual core Celeron Atom parts are almost certainly hitting a more realistic 6-7w at most.
These are not targetting Atom at all. They are maybe targetting Broadwell ULT, but I suspect not really on performance.
azazel1024 - Tuesday, May 12, 2015 - link
To further my thought, last I checked, the Puma design is not significantly better IPC than Silvermont cores...and the dual core Celeron desktop/notebook designs are hitting 40+% higher clock speeds than the Puma+ dual core part...at a realistic 30-40% less power consumption.testbug00 - Tuesday, May 12, 2015 - link
Hard to judge AMD's products until we can get power consumption numbers. At least, for the cat cores. Brazos at "18W" ran rings around a "9W" Atoms and used less power.Of course, that was the old atom and the bobcat core, but, the point is, need to see some systems with these in them first!
I expect Intel to have lower power, higher pref/watt and AMD have higher absolute performance. What that will lead to in realistic scenarios where you are racing to sleep depends on the near-idle power usage.
movax2 - Tuesday, May 12, 2015 - link
Agree. AMD makes the best from 28 nm process. Intel has an advantage on 'TDP vs Performance' here with its 14-22 nm nodes.movax2 - Tuesday, May 12, 2015 - link
In my opinion 10-W quad-core Carrizo-L would make more sense. Clock can be pretty low but with agressive Turbo it will be actually good solution.MrSpadge - Tuesday, May 12, 2015 - link
Problem: the cat cores never had good and agressive turbo. Which is why they either have "large" TDPs or low clock speeds, but nothing like Silvermonts turbo giving +50% single threaded performance.movax2 - Wednesday, May 13, 2015 - link
Actually Mullins (Puma) cores has agressive Turbo. 1.2 GHz -> 2.2 GHz. And that's all for SDP ~ 2.8 W. TDP 4.5 W.silverblue - Tuesday, May 12, 2015 - link
Well, SDP != TDP. As always, it will be power consumption tests that will tell all.Kalelovil - Tuesday, May 12, 2015 - link
@Ian Cutress"Carrizo-L features ‘Puma+’, which by virtue of the naming scheme suggests an updated version of Puma which was seen in Beema."
Beema/Mullins already had Puma+ cores.
http://www.anandtech.com/show/7974/amd-beema-mulli...
I highly suspect Carrizo-L is just a repackaging of the Beema/Mullins SoC in a new socket.
Visual - Wednesday, May 13, 2015 - link
So, can someone tell me in simple words what this all means for me?Specifically, I'm ignoring anything Atom and even Core M, Y etc due to pathetic GPUs. I do not want a step back from the Intel HD 4400 of my Surface Pro 2. Now that Broadwell is out, I guess the new reasonable minimum, or the thing to beat for AMD is the HD 5500.
Can AMD offer better GPU performance in a similar power envelope? And will someone finally make a decent tablet with their part?
sonicmerlin - Wednesday, May 13, 2015 - link
The power consumption of these chips is too high to put in a tablet.Visual - Thursday, May 14, 2015 - link
Sorry, but no, 15W TDP is not too high.I even mentioned a tablet that has similar CPUs in it - Surface Pro 2. Pro 3 also has variants with i5 with a HD 4400, despite also having weaker i3 variants.
Aseda1011 - Wednesday, May 13, 2015 - link
Just realized a ps4 is 2 E2 7110 but with 9 times more shaders 1152/9=128medi03 - Friday, May 15, 2015 - link
I hope there will be at least one notebook model out there with this APU and IPS screen...