
Don't advertise malware writers
Could I ask the f*?!#$%s at The Register not to advertise people who threaten to spread malware. It might be okay to say in general that someone talked about it but this article gave a name.
23 publicly visible posts • joined 14 Dec 2012
I can't see there being any simple solution to Spectre without changes to the protection model. One could run managed code in its own address space but that's expensive. Perhaps better would be having a data protection mask that could be set and which would be associated with any data so managed code could only access what it was entitled to and the hardware could treat it the same way as Meltdown. I can't see them liking anything like that in their page tables though. And it wouldn't be a short term fix.
If the French want to stop seeing these things from google.com they should ban google.com from being accessed by any of their providers, and that's what Google should say to them to do. French citizens can get their government approved information from google.fr
There's no way Google can censor its American operations for the sake of a European country. If the EU wants censorship like Russia and China then that is their own matter and I'm sure Google will try to fully support them in their censorship in their own territories.
That's awful. I was the treasurer for a hall for some years and generated emails or letters and envelopes from the spreadsheet automatically. And I've a speadsheet with CDs in it for a club and collect statistics about popularity. And I've got macros for importing from an SQL database dump automatically including reading the database description and doing some database operations. I think I should keep a low profile or some BOFH is going to open the lift doors for me when when there's no lift behind.
They probably added the last two lines which add up to $500 without extending the range in the SUM rather than having a totals for a table. They were probably added to make the holiday look snazzier or the display fit better. Most of the time Excel will extend ranges properly but there's ways of doing it so it doesn't think you've extended a range. Yep it is fairly intelligent but not super intelligent and some marketing bloke can easily pervert it.
I just had a look at the iPhone 5C video ad on the Apple uk site
http://www.apple.com/uk/iphone-5c/videos/#video-product
The phone looks good but the video just reminded me too much of the video on
http://www.digitaltrends.com/mobile/nestle-mocks-apple-in-kitkat-4-4-ad/
One fifth the size of an Atom makes it something like 10 million transistors I believe which is bigger than the ARM A7 cores - which they consider as far too large for the sort of thing Intel was talking about, When ARM talk about a small chip they mean the ARM M series going from 30,000 transistors up. And there's lots of even smaller ones if you're willing to go for 8 or 16 bit processors. You don't need a PC when you have a processor inside a pill you swalow.
The problem for Microsoft with Intel based machines is that Intel get a load of money, so the extra cost to a customer for Windows compared to other systems is greater than just the Microsoft licence. However they seem to have thought that competing with Android on Apps was the way forward. Well it is to some extent but they didn't push and exploit their greatest strength - that many important programs only run on Windows and the developers want an easy time. The interface to Windows is their strength there, There is a ready market for people to buy Windows RT so they can run packages from the Microsoft ecosystem. That is also a strength of Intel in that people can run Android on Intel machines and switch back and forth to Windows but that isn't a sure way to grow the market for Microsoft or get a bigger part of the money.
The very top item on the Hot Chips 25 page you pointed at had as its first item Tutorial 1 "Heterogeneous System Architecture (HSA): Overview and Implementation" with AMD, Qualcomm and ARM presenters and on Monday there looks to be a talk about a 50% saving of power with some ARM chip.
I must admit I'm rather surprised that Microsoft made so many of them. It should have viewed the RT as a first step in checking what is really needed and to start building up the ecosystem, in fact that's what I thought they were doing with the price. You can't sell something like that easily if even email has problems. They seem to have thought thought they had a mainstream product already done. I'd be sorry if they abandon it because of this experience, I guess it would make their software able to run on things like IBM Power or MIPS as well not just ARM and who can say who will really be still standing after the server wars end with most of the Linux software made architecture neutral?
Those compiler tests mainly just checked if the compilers had implemented vectorization in 2010. That has gone into the latest GCC and LLVM and more is being put in. I'd guess ICC will always be a bit better but nowhere near by the margins shown there. As to vectorization other things being equal the Intel chips should always do better than ARM as the vector register width is 256 not 128 bits. I don't think that'll worry ARM much in the server market but it matters for games and graphics, I guess they're depending on having more cores or using the GPU for such stuff.
Intel must have run a number of tests and then chosen this one to blat about. I was wondering how the hell they got their memory performance figures so it must have occurred to them too that something funny was happening.The test was not working properly with their compiler. It shows a good compiler but exploiting a bug in the test to say things like they did was wrong and just sucks.
There's no point Intel licencing its CPU designs, who'd want them? Intel itself would be too much competition. And it would be attacking its cash cow. Their concentration on the CPU makes them think ARM is the competition. What it should have done long ago is open its foundries and deter the other foundries from competing at the high end. However its foundry plans looks like it is just trying to use unused capacity. Doing it properly would take a couple of years even if they really wanted to and anything can happen by then - I think the technology gap is closing as the foundries all come against some hard limits.
I'm impressed they thought to keep statistics and crunch the numbers. I'm a bit surprised, I'd have thought there would be some correlation but I also remember a study which said the best predictors of software design performance were how quiet the environment was and how big the desks were and whether the had space to store things and that the software tools were a very minor factor.
I hope they go on to the next level now and consider more how teams work rather than concentrating on people. I've always believed it is best to have a mix of people who have different skills - you always need for instance a person who keeps things organised or one who gets people working together and I'm sad too often good teams are broken up and the people moved around.Anyway it really needs the capability of a place like Google to test that and see if that is a better predictor than this concentration on individuals or if i'm just wrong on this.
Those spam messages with the words spelled wrong - they're messages from terrorist leaders sent out so no one knows who is their cells! Many of those pictures of cats posted as well, what the Lolcats do is really a secret message. They're not funny at all. And when people trash a comments page with puerile rubbish - that can be them hiding their evil work too, even young children seem to have been recruited into their ranks. I'm all for the NSA and GCHQ securely storing all these secret messages, of course I advocate using the power of the law to convict the villains but I'll understand if in the interest of national security and if they can't fully decrypt the messages they deal with this menace quietly and without fuss - if some are caught up in this who aren't terrorists I'm sorry but they should not have defaced a web page and anyway you can't make an omlette without breaking eggs.
That's rather an expensive facility for the top end and not exactly clean yet. According to http://researcher.ibm.com/researcher/files/us-pengwu/BGQPerfPaper-final-PACT12.pdf AMD have been looking at it just not announced anything. You can see there the various problems and opportunities at the moment and a comparison with the software equivalents. I think it is also interesting in that it may improve straight line high performance computing in the future as it is the same sort of hardware as is needed for speculative execution which allows for a longer pipeline and yet might give another 10-20% performance boost.
That's not quite right. The x86 architecture needs a lot of memory interlocks like checking for writing into the code being executed and has much stronger coherency requirements besides being saddled with all sorts of strange operations. AMD's general manager of the server business unit said it took them more than ten times more money as twice as long to design an x86 chip than an ARM one. That's down to all the messing around and it'll tell now that the whole business of designing is getting more standard but the actual designs are getting more complicated
Those figures are interesting though as others said one can't say anything definite as no real details are provided but the one that struck me most was them saying the Intel chip had four times better memory performance if I read it right. That would probably explain most anything else and I'd really like to know how it was achieved.
I'm fairly sure Intel will be able to cream some of the high end market off when they get their new mobile chips out whatever about whether these figures one way or the other. Personally though I think the more worrisome strategy of Intel as far as ARM is concerned is that it is trying to get a better presence in the chip foundry business. If they could knock out the high end of the other foundries they could then start causing real trouble for high end competitors and properly protect that market together with their high margins which all this sort of work doesn't really.
The PCs seem a bit expensive for that use too. But that report in PDF with its endless list of acronyms, tables dates, costs, footnotes...., yes I now see where all the money went! I could easily have fixed the PCs for far less and made a very tidy profit but I could never have produced such an official looking report. Yes thats the sort of report you send out to people and require them to read and memorize and then they will be very very careful never to do it again!
Net removal zero is far better than adding to the total which is what things like digging up coal or fracking do, natural processes remove the stuff eventually. Anyway I'm glad to see the Register publishing something that mentions climate change that isn't complete bollocks.
Is God going to stop global warming? I don't think he will anymore than He stops people getting cancer when they smoke. You need to show what's wrong with the science.
As to the ice did you read a bit further about why there might be more ice? That more fresh water in the top layer means it freezes more easily. Like to make a guess where did the extra fresh water in the water around the Antarctic come from?
So now the the Register says the IPCC is putting out the same message so it is boring and anything in the report should be ignored. That 'adapation' idea too, is that in any way different from saying our children will just have to live through all the consequences that have been predicted by the IPCC if we don't do anything?