Carrizo-L is 28nm? What the heck...
Well I hope the 20nm Skybridge APUs aren't cancelled, perhaps we'll see Puma+ on 20nm in March 2015
Carrizo-L is 28nm? What the heck...
Well I hope the 20nm Skybridge APUs aren't cancelled, perhaps we'll see Puma+ on 20nm in March 2015
20nm APUs coming in Q3 2015 per Digitimes.
I was comparing to that to show that while something is twice as powerful in terms of graphics performance it basically receives the same score on that subject. 66-68%
It is bandwidth bottlenecked why do you think low end solutions from Nvidia come with GDDR5. Higher latency higher bandwidth GDDR5 removes congestions and increases the efficiency over DDR3 the energy consumption per unit of performance will go down.
Nobody uses the Yogo pro unless they are complete fools. It is a crappy made products with a crappy made chip. "a power hungry screen"
proof please :|
I'm quite sure pushing those pixels consumes more than the actual screen backlight and tcon combined.
yeah
This benchmark is a terrible pick and I'm not sure if you picked it on purpose to make it seem like Kaveri is better than it is.
Check the hardware they used.
6790K...not 6800K.
At first you might think "but hey, that's just 100MHZ"...wrong.
The 6790K in that benchmark was powered by 1866mhz ram because that is what it officially supports while the 7850K was powered by 2133 also:
The 6800K supports 2133mhz Ram AND is clocked 100mhz higher than the 6790K
The 7850K would have lost vs a 6800K in a CPU performance test. (Which also means 760K > 860K in raw CPU performance
I'm not sure if you're trying to spread misinformation on purpose or if you actually didn't research before posting this. There is a performance difference between 6790K and 6800K...enough to invalidate your entire post as it actually beats Kaveri in CPU related performance.
Kaveri still has reasons why its superior overall (Namely iGPU and PCI-E 3.0)...but still...don't try to bullshit a bullshitter
Carrizo-L is 28nm? What the heck...
Well I hope the 20nm Skybridge APUs aren't cancelled, perhaps we'll see Puma+ on 20nm in March 2015
In this case it doesn't make much of a difference. Carrizo is made with HDL...so the chip still GAINS 0.7 BILLION transistors with 28nm and same die size.
Don't expect wonders, though...those transistors are all going into SoC and utility XD.
Carrizo is an oddball....#dealwithit
"Carrizo" and "Carrizo-L," are scheduled to ship in 1H 2015, with laptop and All-in-One systems based on the 2015 AMD Mobile APU family expected in market by mid-year 2015.
Now you are changing the question. My point was 5x is wrong, its more like ~2x (often less). As I said before NBC subjectively evaluates their notebooks with regards to form factor, market and price. Their ratings are not subjective and should not be compared as such (no one on this forum has attempted to do so that I recall).
As for the power hungry screen.
idle, min brightness, wifi off, low power settings = 3.7W
idle, 200 nits, wifi on, performance settings = 8.9W
idle, max brightness, wifi on, performance settings = 10.7W
wifi and performance settings do not cost 7W
Low end nvidia solutions on the laptop market do not come with GDDR5. Nvidia loves pushing DDR3 (on the 850m or 750m) as do AMD for price reasons. You can push 750m performance (DDR3 version) with DDR3, several times more than beema.
5.4 Wh (~8 h of use) idle, Power Saving Mode, screen at 0%, Wi-Fi ON, keyboards back-lightning OFF;
7 Wh (~6 h of use) very light browsing and text editing in Google Drive, Balanced Mode, screen at 70%, Wi-Fi ON, keyboards back-lightning ON;
Ok not 5 times at much but isn't the difference between playable and not playable what matters. Twice as slow and therefore not capable of running anything is just as bad as 5x as slow and not able to run anything. Though you're correct Intel's horrible solution is less horrible than I anticipated. Gain a 66% for a product that can't play gamesNow you are changing the question. My point was 5x is wrong, its more like ~2x (often less). As I said before NBC subjectively evaluates their notebooks with regards to form factor, market and price. Their ratings are not subjective and should not be compared as such (no one on this forum has attempted to do so that I recall).
As for the power hungry screen.
idle, min brightness, wifi off, low power settings = 3.7W
idle, 200 nits, wifi on, performance settings = 8.9W
idle, max brightness, wifi on, performance settings = 10.7W
wifi and performance settings do not cost 7W
Low end nvidia solutions on the laptop market do not come with GDDR5. Nvidia loves pushing DDR3 (on the 850m or 750m) as do AMD for price reasons. You can push 750m performance (DDR3 version) with DDR3, several times more than beema.
That 2500MHz was meant as effective double rate so indeed 1250MHz. They have them in modules and these dedicated once can go higher. Plenty of DDR-2666 available.I think its also that AMD isn't using the cache and so needs more bandwidth. Intel's HD 4600 on mobile generally matches or is a couple percent slower than the a10-4600m even though it has the same amount of bandwidth. Let alone slower trinity/righland models.
Need to remember that mobile kaveri scales with cores and speed. A 256 shader low clocked model is much slower than a 384 higher clocked model even with the same 1600 ram.
Generally for DDR3 is 900 mhz. I have never seen 2500 mhz (1250 mhz per channel) DDR3 at stock on any notebook.
Your M255 runs at 940 mhz according to NBC, unless you are substantially overclocking the kaveri chip they are not at the same clocks. It is not by any means better than the 750m which is roughly twice as fast.
That 2500MHz was meant as effective double rate so indeed 1250MHz. They have them in modules and these dedicated once can go higher. Plenty of DDR-2666 available.
Well no it doesn't run at 940MHz and my FX-7500 doesn't run at 553 either. But I can OC my Kaveri igp with the AMD overdrive utility to match the clocks of the R7 M255. The R7's performance is better in AIDA's benchmarks by a lot 30-40% at the same clocks. :thumbsdown:
I figured it should ran higher effective clocks than the highest DDR3 available for the whole system.How about nobody uses DDR3 RAM that fast on mobile GPUs. The fastest I've seen at stock speeds is 1000 mhz. Doesn't matter if its possible, I've not seen it and unless you can dig something up dgpus on mobile simply don't use 1250 mhz DDR3.
AMD lists it as up to 940 mhz. That would be the R5 m255 as I can't find an R7 m255.
http://www.amd.com/en-us/products/graphics/notebook/r5-m200
What notebook is this anyway?
Anyway, it looks like there is no HBM. So no solution to bandwidth yet.
65W TDP is literally the only benefit you will see on the CPU side (gaming performance of the CPU may take yet another hit down due to lower MHZ and even smaller cache)
It is possible actually we saw the Gardenia markings 5 months ago the exact same board text showed up in a benchmark suite called Geekbench. It had 1MB L2 per core and was just a small core part just like Beema.Ok, now that we know Carrizo and Carrizo-L will be different CPUs, is there any indication that Carrizo itself will have its cache cut? I thought the mentions of reduced cache were in regards to Carrizo L, which everyone rationally assumed would be a low power variant of the same CPU core as Carrizo--since this is not the case, might Carrizo still have the full cache with Carrizo-L having half cache, just like Kaveri vs. Beema?
It is possible actually we saw the Gardenia markings 5 months ago the exact same board text showed up in a benchmark suite called Geekbench. It had 1MB L2 per core and was just a small core part just like Beema.
I think full Carrizo could still have all the cache in place but only time will tell. (or a nice leak )
Ok, now that we know Carrizo and Carrizo-L will be different CPUs, is there any indication that Carrizo itself will have its cache cut? I thought the mentions of reduced cache were in regards to Carrizo L, which everyone rationally assumed would be a low power variant of the same CPU core as Carrizo--since this is not the case, might Carrizo still have the full cache with Carrizo-L having half cache, just like Kaveri vs. Beema?
You can't even buy Beema or Mullins in notebooks, so why bother? They have been "released" for months and are still available in only one or two notebooks. Carrizo will be "released" next year February and you will be able to actually purchase it in a functioning notebook around 2017.
Wow. That's way too late. Why don't they just launch a successor to Mullins with Puma+?
:hmm: 'muricaYou can't even buy Beema or Mullins in notebooks, so why bother? They have been "released" for months and are still available in only one or two notebooks. Carrizo will be "released" next year February and you will be able to actually purchase it in a functioning notebook around 2017.
It sure seems that way.It really is a mystery where all those Beema/mullins parts went. They must have produced several hundred thousand at least and it's like they've just vanished. Maybe all to Asia and Europe