Search results

While complicated by the turbo behaviour of more recent cards, it would be really interesting to see a clock-for-clock comparison across the various GCN versions.

Anyone remember the 8800 Ultra launch, with Unified Shaders (now used by every GPU)? Or the good old GeForce 256 (hello there hardware texture & lighting)? I'm more surprised so many engines have raytracing support in place already (though Unity appears to be noticeably absent). Then again...

I'd like to see the text around those exceptionally small snippets. With every contract I've read, I could make them say about 5 mutually opposing things depending on how I cropped bits out of them.

The accusation that Nvidia are offering brand promotion only to brands that are exclusively for Nvidia GPUs: well, duh? That's rather the entire point of the exercise. Note that even by Kyle's admission there is NO clause preventing the same company selling cards from both GPU manufacturers...

Early 2019, and that's for Radeon Instinct. Consumer stuff will follow sometime after that. I kind of suspect Navi may come simultaneously with 'Vega refresh', unless the cost of both HBM2 and interposer production and integration drops dramatically in the next year. AMD may want to cut their...

Immune, he says! TEMPEST may prevent RF leakage (unless you duct it through the building;s plumbing as a waveguide) but everyone needs power. I picked that link because it's a passive technique requiring no modification of the target, but if you can infiltrate code you can exfiltrate data via...

So we've got: UHD 120Hz G-sync HDR... but only 27" (and hit by panel delays) 32" UHD G-sync... but no HDR and only up to 60Hz 35" 240 Hz Ultrawide G-sync HDR... but delayed yet again due to lack of panels (and no UHD) UHD 120Hz G-Sync HDR ... but 65"! (and probably costing both arms and legs)...

If you're looking to buy a Rift and meet the recommended spec, you're good. Oculus mandates anything on Oculus Home (outside of the Early Access section) be able to run on that spec, so even if you buy the same game from Steam those optimisations will still be present. This will be true for the...

Generally known as "Van-Eck Phreaking". It;s one of the easier emission attacks, as VGA is nicely analog squirted over a long antenna (the cable) and already helpfully encoded onto carrier signals. The came attack, with more sensitive detectors and more decoding workload, and be performed...

MELTDOWN also affects some ARM cores (including Apple's custom variants), and Redhat's listing for CVE-2017-5754 also covers POWER variants so that may be vulnerable too (still looking into this, IBM's statement doe snot specify which vulnerabilities they are affected by, just that they are...

Looks like these are US only for now (no resellers, they only ship to the US when purchased directly), which is unfortunate.

It's a sounding rocket, rather than an orbital rocket. There are a huge variety commercially available, so this is more surprising that Japan does not have a domestic sounding rocket supplier than anything else.

The Connection Machines devices had a novel system architecture, but were not magically AIs because of it, general or otherwise. There is a lot more to it than mere architecture.

I can see two ways of 'general AI' coming into existence: 1) A dramatic change in our level of understanding at a fundamental mechanical level of what consciousness is and how it works, and the ability to replicate it artificially and arbitrarily or 2) An evolutionary development from...

I would hope these are not just standard RFID tags like those used in pet tagging (i.e. that do little more than yell thie ID number when powered by an external field), but proper encrypted tags using D-H key exchange to authenticate, similar to NFC cards used in public transport and access...

Having some background in Machine Learning[1], I'd have to err on the side of Zuckerberg here. Elon Musk is indisputably an excellent engineer, but from what he has said publicly on AI he has a laymans understanding of the field. The 'existential' worries around 'advanced' AI are vastly...

For a pure IMU-based suit, they;'e already undercut by Noitom's Perception Neuron, which is $1500 for a 32-IMU (vs. 19) array.

We've got bits of the reason why in this thread, but not the whole story: Most* OLED panels utilise a rolling update, usually 5 segments in sequence. Conventionally this will be top-to-bottom in their intended orientation, vertical for phones. Some will be side-to-side if they were originally...

<fatherted>The chip here is small, but the busses between chips there are far away</fatherted> The femtosecond timing used here is because the timings are pre-set, you take many, many, many femotoseconds to tell all the elements "OK, you log the magnitude at this point in the phase of the...

I'm running a 1080Ti + 6700k on a Silverstone ST45SF (older and less efficient SFX SPU) without issue. The SF450 can easily handle any single GPU currently on the market unless you start pushing OCs.

And now Valve's end-game is revealed: the ability to virtually wear your virtual hats (that you paid for with decidedly non-virtual cash). Billions, they'll make, billions!

It may be worth contacting Silverstone: Even at absolute peak momentary draw the Titan X(p)s draw 250W each (500W total), so the remaining 300W should be more than enough for the Xeon + mobo. It could possibly be a current slew rate issue, but that would be a "hey Silverstone, Y U no hit ATX...

Intranet of things is what I want. I very much want my devices to be able to talk to each other on their own private vlan, but I can think of few if any reasons for them to have any access to the outside world.

The interesting thing is the use of CoTS parts rather than bespoke boards (e.g. Zotac's Magnus line). In layout, it's very close to an upended Dan A4, with AIO radiators slapped onto the sides rather than using air cooling. I'm not a huge fan of the aesthetics (though there are far worse...

If this heatsink is targeted for 'general use' rather than just for cases with a low Z-height and a flex riser, then you are now competing with a huge number of existing heatsinks with very similar designs. The AXP-100, NH-L12, Shadow Rock LP, Shuriken, Kozuti, Kodati, and so on. The heatsink as...

I though it was fairly well established that the TIM used is just fine, it's the separation between the top of the die and the underside of the IHS that is the issue. Which is why relidding is so dependant of proper seating and application of pressure (and why convex heatsink bases perform...

The TIM in Ivy Bridge and onward is just fine, it's the clearance gap between the die and the IHS that causes the 'issue': A gap is critical to prevent damage to the die when force is applied to the IHS. The smaller the gap, the less force can be applied before the IHS contacts the die and...

'Virtual Retinal Displays' using direct laser projection currently do not exist outside of lab prototypes, and are limited to low fields of view. The Avegant thing however is pure grade-A marketing wank. What they have is a bog-standard LCOS microdisplay setup (same as the setup using a DMD...

If the Ryzen dies are indeed smaller (or even the same saize) as 4C8T Ivy Bridge or later, then that would rule out a soldered IHS. It will be interesting to see how close AMD are willing to place the IHS bottom to the die top: Intel's 0.06mm is conservative, trading off thermal performance for...

The problem with the idea of an 'AR contact lens' is that you cannot focus on an image placed on your eyeball. Because of this, any AR contact lens would itself need to contain extremely powerful thin-film optics to allow you to focus on a close to 0 focal distance. Such optics do not currently...

UK here, DHL tracking is still listed as awaiting collection, though DHL could just be slow in updating the status. ::EDIT: Now marked as collected!

Hey Saper, Zaber. The Cooltek LP53 with the NF-A9x14 is an attractive combo. The LP53 is a bit tricky to get in the EU, but Alternate have them in stock at the moment for a reasonable price. I've got one on order at the moment, so if you don't want to splash out on yet another heatsink to test...

That is the LGA 3647 socket assembly compared to an LGA 2011 CPU. The Dynatron B2 will cover the area bounded by the outside edge of the rectangular retention bracket around the socket. And it has a completely flat base,so there's no clearance below it unlike the LGA 115x stock HSF. The LGA 115x...

The Dynatron B2 is an LGA3647 heatsink. The socket is almost as long as a full (not SODIMM) DDR4 DIMM! I don't think it's going to fit onto any LGA 155x board due to clearance issues with components around the socket. You'd need to add a shim the same thickness of the highest component covered...

Onboard PCIe slots DO use shielded paths, the traces running inside the motherboard layers are shielded by being surrounded by ground planes (stripline). This is done both to shield the PCIe lines from all the EMF producing parts on the motherboard (power phases, other buses line USB3, etc) and...

Use a latency tester (e.g. the one that was released for the DK1) to measure unidirectional latency. Or place two HMD side-by-side using a HDMI splitter to compare wired vs. wireless with a photodiode timer, or even a high-speed camera. Or even just have a wired HMD and wireless HMD next to each...

Opened the page again, and the timer is now showing the correct time (ending at 1700 UTC). I am on the GMT line (I can actually look out the window and see the Greenwich Observatory!), so my timezone is currently UTC.

Hi SaperPL / ZombiPL , the newsletter has the launch time as "18:00 GMT+1" (1700 UTC), but the countdown timer on the website is currently due to hit 0 at 1600 UTC.

If you want a flat air-blower, then the Murata Microblower line may be suitable (though it emits axially rather than radially). If you want a radial piezoelectric blower and can tolerate bit being a bit thicker, there's the Synjet blowers

I have the Z170 Fatal1ty, along with a Rift and 4x sensors. Works just fine, I have had no issues with it. Sensors and HMD are all on the Intel-provided USB 3 ports (not the 3.1 ports from the ASMedia controller), with 2x sensors on USB3 and 2x on USB2 via the provided extension cables.