this is pretty simple, the 17W part will most likely keep higher clocks for longer while the 19W part will throttle more.
Unless the A6-7000 is just really bad silicon maybe. It probably is a salvaged die that has the other parts still physically there, and I think some of those parts even if not in use can still take energy. Also, for the A6-7000 to have a lower speed and half the cores on the GPU/CPU, it must really be the 'worst' part from the finished wafer other than completely unusable parts.
*edit - I don't think it is so much that the FX-7500 needs only two watts more for doubling up everything and faster clocks so much as the silicon in the A6-7000 needs the 17 watt TDP just to run as it does due to manufacturing defects.
The AMD AMDs have a larger iGPU than the corresponding Intel Haswell U-series. Also, the TDP is measured with both the iGPU and CPU cores active.
So for CPU intensive tasks where the iGPU is not used much, could it be that the actual power consumption of the 17/19W AMD APUs is comparable to the 15W Intel Haswell U-series chips?
They're subjected to the same radiation in day that the average human on Earth would receive in a year.
I dont believe there is that much radiation if at all in that orbit, otherwise you wouldnt have humans stay for long periods of time in the ISS.
It is likely the exact opposite. The igp kaveri has is vastly more efficient. Look at beema/mullins.
You are comparing 22 nm FinFET vs. 28 nm planar. Haswell has the integrated IVR and PCH.
IVB/HW ULV CPU is tremendously efficient. The igp is the main power user.
Looking at notebookcheck results a 15W tdp fits 2 HT core running prime95 anywhere from 2.2-2.6 ghz. That includes the PCH.
The benefit of SOI is less current for the same performance. The issue in particular is with PD-SOI which by the 45nm needed more voltage to hit the same performance. That issue grew with PD-SOI when shrunk was caused by the floating body which sapped off most of benefits of having SOI. PD-SOI's performance isn't relative to FD-SOI's performance, so you can't compare PD-SOI to FD-SOI unless you are only comparing SOI wafer costs. PDSOI has thick insulator while FDSOI has thin insulator, the cost with FDSOI optimized wafers is lower than PDSOI. While the cost of transportation of wafers for FDSOI is higher than PDSOI.So SOI wasnt the POS every usual suspect here claimed it was? Good to know
Jaguar is substantially better performance than Bobcat. In fact it has higher performance than silvermont both in performance and IPC.
Jaguar is faster Clock to Clock than Bobcat, so a dual core 1GHz Jaguar is faster than dual core 1GHz Bobcat.No, the IGP was a bit better, but thats all, dual jaguars where even worse(1ghz ones), quad jaguars was better because of more cores.
Jaguar is faster Clock to Clock than Bobcat, so a dual core 1GHz Jaguar is faster than dual core 1GHz Bobcat.
No, the IGP was a bit better, but thats all, dual jaguars where even worse(1ghz ones) than dual E-350 bobcats, quad jaguars was better because of more cores.
Im sorry, but oems stopped using E-350 in favor of A4-1250 in 11.6" models, that resulted in abysmal performance.
And by the time Jaguar was launched, C-50 was not longer used either, both the C-60 and C-70 had turbo enabled and it was kicking in when used in 14" laptops, thats why Lenovo did not wanted to "upgrade" to jaguar on their cheap models and stick with C-60 and C-70.
There is a reason of why Celeron ULV became popular on low end and Bobcat was still used after Jaguar launch, and its not because Intel was paying everybody.
Interesting. So the main advantage of A4-1250 over C60 is that they turned it into an SoC, saving ~5W.E-350 was Dual core 1.6GHz 18W TDP + FCH A50M (5.9W TDP) resulted in more than 22W TDP package.
C-60/70 are dual cores 1GHz base with 1.3GHz turbo at 9W TDP + FCH A68M (4.7W TDP) resulting in more than 13W TDP package.
A4-1250 is a dual core 1GHz 8W TDP SoC, no need for FCH chip. CPU performance was comparable to C60/70.
A4-1350 is a Quad-core 1GHz 8W TDP SoC and A4-1450 is a Quad-Core 1GHz base with 1.4GHz Turbo 8W TDP SOC.
Dual core Jaguar A4-1250 didnt replace E-350 but C60/70.
The successor to the A4-1250 is the E1 Micro-6200THow about Beema/Mullins? The successor to A4-1250 should be E1-6010, right? Which is 1.35GHz/10W (0.35GHz upgrade, 2W higher), so this time they're actually aiming for improved performance instead of better power?
for me -even though multithreaded s/w support is spotty at best- give me 16-32 puma+ cores vs 4 steamroller modules+tonnes of cache. Why? if mantle/dx12 is the way forward then fast cpus wont be the end all be all as they are with the current driver module. As for general software, more and more things get multithreaded, also just multitasking will benefit.
example scenario:
playing a game, recording it locally, encoding and streaming in online, multiple browser tabs open on a 2nd or 3rd monitor...not saying this couldnt be done efficiently now with a fx8350 but 16-32 puma+ cores @3ghz would do just fine, that is if it is priced at ~$200
E-350 was Dual core 1.6GHz 18W TDP + FCH A50M (5.9W TDP) resulted in more than 22W TDP package.
C-60/70 are dual cores 1GHz base with 1.3GHz turbo at 9W TDP + FCH A68M (4.7W TDP) resulting in more than 13W TDP package.
A4-1250 is a dual core 1GHz 8W TDP SoC, no need for FCH chip. CPU performance was comparable to C60/70.
A4-1350 is a Quad-core 1GHz 8W TDP SoC and A4-1450 is a Quad-Core 1GHz base with 1.4GHz Turbo 8W TDP SOC.
Dual core Jaguar A4-1250 didnt replace E-350 but C60/70.
ah, so you'd be willing to trade the ability to pleasurably play games now for the possibility of having mediocre to subpar performance years from now
who says a 3ghz puma+ core can't "pleasurably" games now?
and secondly have you seen mantle? I don't see how you can conclude that games will run subpar on such a theoretical apu? Mantle -and by extension dx12- will alleviate many bottelnecks in the graphics pipeline and allow for the possibility of better multithreading, which the MOAR COARS strategy can take advantage of.
so yeah I would sacrifice single threaded perf for more cores and better multitasking and multithreading!
I'm sure the answer's yes, but in the sense that even the processors we use will have errors on occasion due to natural background radiation. Of course, the chance is exceedingly rare...Nice, is that enough to make 32nm and bellow IC malfunction ??
This is AMD we're talking about. They've done stuff like this before. The "Ruby" video back when they announced Bulldozer (I think?) was really embarrassing."Out of this world (or atmosphere at least) performance", or maybe just "out of touch". They wouldnt seriously use such a juvenile, meaningless gimmick to announce a product would they?
"Out of this world (or atmosphere at least) performance", or maybe just "out of touch". They wouldnt seriously use such a juvenile, meaningless gimmick to announce a product would they?