Search results

Yes, I know Renesas :) (They used to be NEC ages ago, so they have had a *long* relationship with PowerVR). Though I personally don't work with them much, being mostly focused on Android right now. I do not know how open to small quantities that we would be interested in, but possibly worth...

Automotive tends to have slightly different use cases - for example the jacinto6 (effectively the progression of the omap5 into the car space) has a very large number of video decoders, but no faster cpu or gpu than the omap5 itself. And power constraints are often a little relaxed compared to...

Yeah, I'm not trying to say "people shouldn't be encouraging public docs/source release", but the "Bad, IMG suxxx!" comments are exactly the opposite way of doing it, being not remotely constructive and only possibly harmful (as in I find it insulting, and you're less likely to get someone who...

I believe TI use 99% of the code from our 'reference' driver, with most of the changes being in platform integration code and whatnot. I would be surprised if there were many (any?) people at TI who knew the 'core' shader compiler or GLES state tracker well enough to make effective changes...

I know I'm "too close" to be considered a remotely impartial observer, but I *really* don't understand these comments. Are you moaning about the lack of public documentation about the PVR architectures? If so I can assure you our customers get them. And again, vested interest, but my own...

No, google *do not* provide a licensed version for individual use. By the reading of the published docs, technically any time you modify your own android image you'd need to run CTS and submit the results to google's partner review process in order to get the gapps enabled on that device image...

The problem is that Google keep their gapps/play services closed, and behind a relatively 'strict' license (You MUST pass CTS in order to use them, and you MUST pass CTS in order to use the brand name "Android"). It kinda makes sense, you don't want people making incompatible forks and still use...

But the comparison here is so far off it *doesn't* make sense to compare anything. So far that I consider this well on the side of "Lies" instead of "Misleading marketing". It's not news that general purpose hardware will be beaten by specialized in it's domain. That's why that specialized...

That's more that a GPU is much more efficient at that specific workload than a CPU. Not really anything about the CPU arch itself. So a completely pointless and uninteresting comparison.

Really, the balance has been moving away from "simple" instruction decoders into compressed instructions sets as cache use and memory bandwidth are getting relatively more expensive than complexity of the frontend on the cpu itself (e.g. arm THUMB) - and x86 could be considered a 'compressed'...

Qualcomm's 'Quick Charge' was a brand name for a method of pulling more power through a USB port than then USB spec allows. They did this with a higher voltage, I believe. The USB type c effectively supersedes this, allowing significantly higher currents to be drawn to provide the equivalent...

Bigger numbers sell, even if it's literally useless, or even causes issues. 24bit is kinda useless for output, as that's realistically more dynamic range than the human ear can detect - though with the current 'loundness war' going on it seems fewer people actaully use the full range, though...

Try building with -O0 (no optimisations) and -g (debug info) - that may make the stacktrace better.

If true, that means the undocked performance will be somewhere between the iPhone 6 and 6s. I wouldn't really want to use that for VR or anything similar, as they *really* need power to get the high frame rates required to avoid sickness.

(Sorry for possible necromancy) Yeah, the SGX driver requires a usermode/kernelmode pair all built using exactly the same versions, even down to the core revision. And I believe that TI were the only people to use a 544SC (single core) - which is slightly different to a 544MP1 (which has the...

I think I internally conflated someone saying 'NVidia using RISC-V' (without the caveat of 'just for their onboard firmware microcontoller') and 'Someone should make a RISC-V chip with a GPU' and ended up with 'Someone should make a GPU out of the RISC-V project', instead of it just being a...

The actual shader cluster is a relatively small part of an efficient graphics pipeline, stuff like texture samplers/geometry processing/rasterisation etc. all work *much* better with fixed function hardware (or semi-programmable, certainly more 'fixed' than a general purpose CPU core running...

For a start: o[object_id].texture_framerate[c] = 10; is going over the array of texture_framerate, it's defined as [8] large, but c >=8 for a number of loops. And "1.4013e-44" is the hex value 0xa when interpreted as a floating point... which is '10' in integer space :)

That's normal if using memory set by an ioctl that valgrind doesn't know about, which is extremely likely when using the mali gpu driver. Valgrind can't trace into the kernel, so it's probably been correctly initialized there, but as valgrind doesn't know what the mali-private ioctls do it spews...

15W tdp. In the mobile world, tegra is a joke for power consumption.