Overclock.net banner

1 - 20 of 75 Posts

·
Premium Member
Joined
·
16,904 Posts
Discussion Starter · #1 ·
Quote:
This brings us to Tegra X1, NVIDIA's latest SoC which was announced today at CES. While largely a continuation of the strategy begun with Tegra K1, there's still a lot of ground to cover. One of the bigger surprises is the CPU configuration. While Tegra K1 had a version with NVIDIA's custom Denver core for the CPU, NVIDIA has elected to use ARM's Cortex A57 and A53 in the Tegra X1. The A57 CPU cluster has 2MB of L2 cache shared across the four cores, with 48KB/32KB L1s (I$+D$) per core. The A53 cluster has 512KB of L2 cache shared by all four cores and 32KB/32KB L1s (I$+D$) per core. NVIDIA representatives stated that this was done for time to market reasons.



For power testing NVIDIA ran Manhattan 1080p (offscreen) with X1's GPU underclocked to match the performance of the A8X at roughly 33fps. Pictured below are the average power consumption (in watts) for the X1 and A8X respectively.



NVIDIA's tools show the X1's GPU averages 1.51W over the run of Manhattan. Meanwhile the A8X's GPU averages 2.67W, over a watt more for otherwise equal performance. This test is especially notable since both SoCs are manufactured on the same TSMC 20nm SoC process, which means that any performance differences between the two devices are solely a function of energy efficiency.
Source

Some interesting details in there. Also NV's first TSMC 20nm design.
 

·
Premium Member
Joined
·
10,262 Posts
Quote:
Originally Posted by kakik09 View Post

Wooooaaaaah. It uses less voltage than the iPad Air 2? So is it at all possible for a watered down version of this to come to a phone?
biggrin.gif
*power
According to the article.

Seems like Denver really got canned btw. Saying it was just the time pressure wouldn't be right because it should've already be out there in Denver version for K1.
 

·
Premium Member
Joined
·
9,327 Posts
Quote:
Originally Posted by jmcosta View Post

lol i just found this

new console coming?
No just stupid PR
 

·
Politically incorrect
Joined
·
9,463 Posts
Quote:
Originally Posted by jmcosta View Post

lol i just found this

new console coming?
nVidia Make slides like that every time they release a new chip. Their PR is as immature sometimes as that of AMD.
 

·
Premium Member
Joined
·
10,262 Posts
Quote:
Originally Posted by sugarhell View Post

No just stupid PR
2 SSM clusters match over a thousand GCN SPs while the 750 Ti matches the GTX480 which matches the 7870. I smell bull. But let's rap up the slide even though it is a fail at hand:
One is x86 based the other is ARM based (arm is more efficient)
One has 8GB of DDR3 and ESRAM consuming tons of power the other does not.
There is a huge efficiency difference between GCN and Maxwell due to a 2 year time gap.
One is fabbed on TSMC 28nm the other on TSMC 20nm.

All with all there are various reasons why this could be the case. But why compare to the Xbox1 and not the PS4. (probably because X(box) one sounds like X1)
They should've named it M1 to be honest.
 

·
Joined
·
2,327 Posts
But can it run Unity?
 

·
Registered
Joined
·
3,472 Posts
no use of overkill GPU performance if the CPU single threaded performance are not good.

So my question is...how good is the CPU really? does it beat A8 single threaded performance?
 

·
Registered
Joined
·
1,796 Posts
Quote:
Originally Posted by maarten12100 View Post

Seems like Denver really got canned btw.
Quote:
Originally Posted by MapRef41N93W View Post

Woah wait Denver just got abandoned? But why? It was actually pretty impressive in the Nexus 9.
Quote:
In any case, much like the still in the dark Parker design, NVIDIA is holding their cards close to their chest and hinting that Denver will still be in future generation products. If Parker is still part of NVIDIA's plans - and they are without a doubt working on some kind of 16nm FF Tegra SoC - then we may still see Denver as part of the original Parker plan. Otherwise at this point while there's no reason not to believe NVIDIA about the fate of future Denver SoCs, it's anyone's guess when we might see it again.
Coming from the AT article.
 

·
Consumerism 101
Joined
·
4,098 Posts
Quote:
Originally Posted by Shadowarez View Post

I think it will run crysis! Time to get a nice tablet with this chip. My Tf700 Tegra 3 is Ancient. 1 Gb of ram.
Already cheap tablets on the market that "run" crysis:
 

·
Consumerism 101
Joined
·
4,098 Posts
Quote:
Originally Posted by Sannakji View Post

They need to get this into a Shield 2 portable and forget about the cars for the time being...
Now why would they do that?
Quote:
Though the mobile device segment is the largest sub-segment in the Tegra division, the automotive segment is the fastest growing and offers higher gross margins (compared to devices). Automotive electronics is a large market and it is going through a transition as cars have increased computing capability in both the drive train and the dashboard. Increasingly, dashboard functionality within cars (infotainment system, digital cluster and automatic driver assistance) are being computerized. Nvidia has been working on building its automotive computing platform for over a decade and is in a strong position to take advantage of this growth. At present, there are more than 6 million cars using NVIDIA processors on the road, and the company expects another 25 million processors to be integrated into vehicles over the next 5 years. Nvidia's client list includes - Honda, Volkswagen, Audi, BMW, Lamborghini and Tesla Motors.
 

·
Registered
Joined
·
3,248 Posts
Quote:
Originally Posted by Clocknut View Post

no use of overkill GPU performance if the CPU single threaded performance are not good.

So my question is...how good is the CPU really? does it beat A8 single threaded performance?
I don't know why it couldn't at least match it, both are using ARMv8...depends on what NVidia did when they were putting it together and if they can streamline performance of the kernel, etc once its in something.
 
1 - 20 of 75 Posts
Top