Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Comment: Re:a 5-year lag (Score 2) 268

by wolrahnaes (#49661589) Attached to: Russian Company Unveils Homegrown PC Chips

You don't need a high end chip for gaming though, it's mostly GPU-bound.

I beg to differ. Some very popular games are also very CPU-bound. GTA V, Kerbal Space Program, and Project CARS are three well known titles I can think of off the top of my head that run a lot worse on the Phenom II X6 1045 (2.7GHz hex-core which I overclocked to 3.5GHz) in my secondary gaming machine than on the i7-4970k (4GHz quad-core, not overclocked) in my main desktop. Both machines have SSDs, 32GB of RAM, and GeForce 970 graphics so the CPU is the only significant difference.

AI and physics simulation still lean heavily on the CPU. Here's some benchmarks from Project CARS (which just came out last week) showing how important CPU performance is when simulating 30 cars on the track: (article is in Polish but numbers don't need translation)

Comment: Re:Not enough logging (Score 1) 199

That's certainly a good idea, though I'd imagine that a lot of the software vendors involved wouldn't bother. I mean the betas for Vista were publicly available for over a year before its release. I ran them intermittently throughout that time and filed bugs or posted on company forums where possible, but most of the responses I got were along the lines of "we don't support beta operating systems, we'll start working on Vista support when it's released".

Vista -> 7 was a mostly painless transition aside from a few apps that stupidly have maximum version checks and refuse to install on a new OS no matter what, but then the same thing happened with Windows 8 and Server 2012. The vendors who tend to cause these problems just don't care. I have more than one vendor right now that still in 2015 insists that we disable UAC. Fortunately we've found that we can install it with UAC disabled and then immediately turn it back on, but this is the level of incompetence we're dealing with.

We're basically stuck in a never-ending cycle of stupid when it comes to specialty business software. Businesses don't upgrade because their vendors manage to do dumb shit that breaks when you upgrade things, and the vendors don't have any incentive to fix it until a lack of availability forces their customers to start upgrading.

Comment: Re:Enterprise Turnover? (Score 1) 199

Surely will. A perfectly working system stops working because Microsoft singlehandledly changes the system but still the blame is for a third party and the solution is me expending more of my hard earned money?


If it was only working because it depended on a bug or internal data structures that it wasn't supposed to be playing with, it wasn't "perfectly working" ever.

I can write a program that does a lot of things horribly wrong but works on Windows XP because it tolerated a lot of bad behaviors, which won't work at all on a more modern system. Is that Microsoft's fault that I wrote it wrong?

How many user-level apps were writing to system directories without reason all over the place in XP and prior which "broke" when Vista stopped letting them do that? Not a single one of those are anyone except the developers' fault.

Comment: Re:Single shop most likely (Score 1) 323

AC is correct, I do not care at all about the embedded key and would happily delete it if possible. I want to use the separate retail Pro key that I have, but the Windows installer insists on using the embedded key that is of no use to me unless I go way out of my way to convince it otherwise.

Comment: Re:Single shop most likely (Score 1) 323

You missed the point.

I'm trying to do a fresh install of Pro. The laptop has an embedded key for non-Pro, which I have absolutely no interest in.

Windows "helpfully" jumps right past the key prompt in the installer when it detects an embedded key. Because Microsoft is Microsoft, for some reason though there is an upgrade facility I can't even give up and just put my key in to upgrade the non-pro install as it's not an upgrade key.

Comment: Re:Single shop most likely (Score 2) 323

The only problem it can sometimes cause is if you're doing a cross version and cross type install without an existing OS on the box (ie it came with 7 home and you're doing an upgrade install of 8.1 Enterprise)

And let me tell you, trying to install 8.1 Pro on a Lenovo that shipped with regular edition 8 is a test of patience. The installer *really* wants to read that key and is not easily convinced to ignore it and let you enter the key that you actually want.

Comment: Re:Two things... (Score 1) 65

by wolrahnaes (#49598945) Attached to: Game:ref's Hardware Solution To Cheating In eSports

Consoles aren't fool-proof. But other than the PS3 there's no easy way to inject arbitrary code. So other than taking advantage of bugs (which are the developer's fault), you can't really cheat on something like the XB1 or PS4 like you can the PC.

Cheating the PC, by comparison, is almost always accomplished via arbitrary code. Wallhacks, aimbots, complex macros, tools that unveil more data than the player is meant to see, etc.

Every single last-gen console was hacked wide open. 360 and Wii will happily run arbitrary code just as well as the PS3. Last time I checked the Wii was still a purely software mod, no hardware required. Xbox 360 requires hardware unless you have an old console that hasn't been updated in years, just like the PS3 now that both have patched their major security holes.

Comment: Two things... (Score 4, Insightful) 65

by wolrahnaes (#49596817) Attached to: Game:ref's Hardware Solution To Cheating In eSports


Consoles are almost completely devoid of cheaters because they provide anti-cheat solutions baked-in their hardware.

I'm not sure what consoles this guy has been playing, but cheating is rampant in pretty much every popular console game. Some kinds of cheats may be harder to implement on consoles, but they always find ways to do it.

Second, all his rig does is monitor USB inputs. The same USB inputs I can fake using literally the same Arduino hardware he seems to be using for his prototypes. Any kind of macro-based cheats would be trivial to implement on USB-capable microcontrollers. One's cheat program of choice just has to change from sending fake inputs directly to the OS over to passing the same input commands out to a simple piece of hardware which then sends them right back as USB HID inputs.

Comment: Re:Requires Line of sight (Score 1) 96

by wolrahnaes (#49523275) Attached to: Optical Tech Can Boost Wi-Fi Systems' Capacity With LEDs

Until 2012, MacBooks have had this built in... perhaps this might be something useful to put in a spec as a NIC option?

The late PowerPC and Intel through 2012 consumer Macs did have an infrared receiver, but it is only a receiver for the sometimes optional, sometimes bundled remote and can not be used for two-way communications. The last Macs to have IrDA were the G3 Powerbook and Bondi Blue iMac. The multicolor iMacs, iBook, and Powerbook G4 all dropped it.

Comment: Re:Thank god (Score 1) 229

Not true, I haven't played TF2 in months but over the last two weeks or so have received a friend invite from a clearly bogus account about once a day. I suspect that's indicative of a major increase in these things because prior to just recently I hadn't had more than a dozen in the many years I've been on Steam.

Comment: Re:Yay (Score 1) 72

by wolrahnaes (#49477723) Attached to: New Samsung SSD 840 EVO Read Performance Fix Coming Later This Month

Not to mention their phones need ODIN which is Windows only. It's a clusterfuck from hell when you go to bed with Samsung and Microsoft.

They don't *need* Odin for anything normal people do, we're not even supposed to have Odin available to the public. That said Heimdall is open source, cross platform, and in my experience works better than Odin. Even on Windows it's a better tool IMO.

Comment: Re:Great (Score 1) 106

by wolrahnaes (#49438927) Attached to: Bell Labs Fighting To Get More Bandwidth Out of Copper

My VDSL2 connection is sold as 100/10, which according to my router is currently connected at 86552/10000 kbps. It's very common around here to have fibre to apartment complex and then VDSL2 over telephone lines for last mile.

My cable connection is sold as 100/10. It connects using an 8x2 DOCSIS 3.0 channel bonding setup, providing a line rate of 343.04/61.44. DOCSIS overhead makes this around 304/54 usable. My modem's config file has it limit me to 115/11.5 and I can generally use every bit of that.

And this is indeed my personal connection, not shared with everyone on the same cable loop.

What sort of a connection do you think is upstream of the DSLAM? I guarantee if even a fraction of the users you share hardware with tried to max out their connections at the same time you'd see the same problems as an over-oversubscribed cable node. There are also plenty of cable providers, mine included, that don't push their oversubscriptions to the limit. I've never had a time where my speed underperformed the rating by more than a few percent where I wasn't having actual errors on the line.

MATH AND ALCOHOL DON'T MIX! Please, don't drink and derive. Mathematicians Against Drunk Deriving