Posts by Steve Todd 2645 publicly visible posts • joined 19 Sep 2007
If only 1 in 10 of your drives fail in a 5 year period then you've blown your profit margin by the time you figure out the cost of replacement devices (even if they are cheaper), staff to process the RMA, postage etc. You've also blown your reputation and are going to struggle to stay in the market.
Why should they? They are only interested in finding malware that runs under iOS. If an app writer wants to distribute Windows malware then simple encryption of the package data will block any virus scanner. They may well scan any mails generated by the software, but how else do you expect a desktop OS to get anywhere near executing a mobile app?
No, taking a bath is when you have to sell an asset for less than you bought it for.
If Apple had a stack of iPhone 4S phones sat in a warehouse that they couldn't sell and had to discount then that would be taking a bath. As it was they kept their supply chain stock at normal levels (they aim for 4-6 weeks of stock) and cut back orders from their suppliers to match.
As people keep trying to tell you, the way that business works is to compare against the equivalent quarter from the previous year. They do this because there is normally variation in sales over the year (peaks at Christmas for example). On that basis Apple didn't do badly (20% up on sales) so this is far from a disaster.
Except if you apply the same rules that the courts do when comparing against a design patent then the 2001 tablets, had they come out after Apple's patent, wouldn't have infringed the design.
Samsung and their lawyers have already dug up as much prior art as they could, and so far the judge isn't buying it. Your blustering has about the same effect.
Enough with the stupid "Apple patented a rectangle with rounded corners" stuff. The courts certainly don't think that. The Xoom, not covered by the patent. The GTab 10.1 covered. The GTab 10.1N not covered. All three are rectangles with rounded corners, and only 1 was found to infringe. Small changes fixed that.
The point about FRAND is that something like 3G is made up of hundreds if not thousands of patents from many sources. If each were to ask for a percentage of RRP like Samsung are demanding then you'd pretty quickly end up with a demand for more than the retail cost of the device. The ND part of FRAND means that you must charge anyone who wants to use the patent the same amount. If EVERYONE has to pay Samsung $15 to use their patents (which would add about $45 to the RRP BTW) then fine, but given that other users get the patents included in the cost of their baseband chips that's unlikely.
1) see my post below, you're not comparing like for like numbers
2) you're saying that less than 52% of Apple's sales are for the latest model? Given its sold as a premium brand that's unlikely.
3) This was a model for model comparison, not how many phones Samsung sell in total. Many of what count for Samsung's total sales are cheap models like the Bada range.
Samsung have a habit of quoting sell in numbers (how many have they sold to dealers and the network operators).
Apple quote sell out (punters taking them home and activating them).
The two numbers are hard to compare as you have no idea how many of samsung's phones are sat in a warehouse or on a shelf. For a worldwide release it can easily be a couple of million.
And you missed the bit about Microsoft making a legal commitment to give users a choice of the browser they want to use when Windows is first used?
Most ordinary users don't realise that they HAVE a choice. Even if they end up using IE they need to be told that there are other options.
You don't seem to understand what a monopoly is. A company doesn't need to own ALL of a market to hold a monopoly position. It needs only to hold enough share to be able to unfairly influence competition.
In Microsoft's case there are alternative operating systems, but people buy software not operating systems. If the software they want is only available on Windows they buy Windows. Because most users have Windows developers write their apps for that. The result is a self-reinforcing loop.
Add to that Microsoft have terms to OEM manufacturers that make it a disadvantage for them to sell a PC without Windows and you end up with the current situation where Microsoft owns 95% of the market.
And that's different from an Android APK file how? (hint, APK files are renamed ZIP files just like Java JAR files)
Neither Android or iOS require the user to have a PC. You can pick an app from the store and have it download and installed directly. Java makes life harder in this case.
Nokia + RIM < 20% of the smartphone market. How does that equate to Java winning the smartphone wars?
Android is only a Java-like platform (and has the results of a court case to prove it)
iOS is based on Objective C/C++
Combined they own the other 80% of the market and look like the winners to me.
Screen resolution is only one component of the whole picture, and probably the easiest to deal with.
The first problem is in software APIs. If you write your code to make use of the latest APIs then you limit the size of your buying market. With Android 90% of users are on version 2.3 or older, therefore most code will be targeted at that version or older.
Second hardware capability. An x86 processor, even at the lower end of the range, is much more capable than an ARM CPU. With Android your users can have anything between a single ARM v11 and four Cortex A9's (or even an Intel Atom these days) so you have to pick a minimum hardware performance level. The range of GPUs in use causes a similar problem (especially with older APIs not giving full GPU acceleration).
Thirdly compatibility testing. Because of the first two points the range of devices you need to verify your software against is much larger. This will take longer and cost more in test devices.
Firstly 63% is the percentage of web accesses, not users. Android users keep telling us how they are selling more mobile devices than Apple.
Secondly its not a monopoly position anyway.
Thirdly (and what got MS into trouble) its not illegal to hold a monopoly, what IS illegal is to use that monopoly to unfairly gain advantage in another market. Forcing users to take IE (to the point of nailing it into the OS and banning OEMs from installing third party browsers) in order to gain market share was what caused the trouble.
that Apple had entered into a legally binding agreement to allow users to pick their browser of choice in iOS in settlement of an anti-trust investigation. Until that happens (and Apple hold only a small fraction of the worlds browser use) you won't see it either.
The standard Surface tablet has an ARM CPU (not something you can throw heavyweight apps at), allows code installs only via the Microsoft app store, will only run Metro apps (no Windows Desktop or legacy code) and can't take part in Active Directory. How does this make it in any way better than the iPad? The keyboard cover? I can't see this being great to type on, no key travel or feel, and you can already use Bluetooth keyboards with the iPad.
The Surface Pro is Intel Core i5 based, it is already known that it will cost more than the surface. It's also known that it's battery isn't hugely bigger than that of the Surface, but the TDP of the ULV Core i5 is 17 watts, roughly 8 times that of the ARM. Tell me how it is going to match the Surface for battery life?
Kind of depends on what you're tring to create. The meme that you can't create on the iPad has been disproved time and time again. Unless you have a Bluetooth keyboard (which are hardly rair or expensive these days) typing isn't much fun, but many other things can be created on it that work rather well with the touch interface.
As for AD, that only applies to the Surface Pro. The regular Surface has no such enterprise features.
If you're trying to do that your design architecture is wrong. It's more like a smart terminal than a PC, and as such you should be offloading heavy work to servers where it belongs. Windows RT is no better, and full fat Windows 8 has a much higher hardware cost combined with lower battery life.
There are many ways in which the iPad can work well in a corporate environment, Citrix for example gives you the ability to access both conventional an tailored Windows apps. Stop trying to think of it as a PC and you'll find it can solve a lot of business problems rather well.
Then you either have a large, not very portable laptop or you don't do work that requires significant amounts of screen real-estate. There are very few laptops that can match a 2560x1440 monitors resolution, non at all that have a 27"+ screen. Not all, but a huge proportion of laptop users at least have a desktop monitor because of this.
Given the complaint was that 1GB ethernet was NEEDED in order to shift data at 100MB/sec (by the time you knock network overheads off you'll be luck to hit 80MB/sec anyway), people who use HUGE files, like video for example, need to be able to archive it faster than that.
Final point, this isn't about what you personally want from a laptop, its about what there is a significant demand for. Given the number of Windows laptops that come with docking ports and port replicators you have to admit that there's a demand.
Did you not read the bit about it giving you extra ports? Your laptop magically gives you a FireWire port when you connect it to a monitor? With Thunderbolt you keep all your existing ports and gain a batch of new ones.
The problem with netbooks is twofold. Firstly crappy screens. Intel limited them to 1024x600 for no apparent reason. Sorry, in this day and age that just doesn't fly.
The second problem is the horribly underpowered Atom CPU combined with Intel graphics, a recipe for poor performance if there ever was one. You really can't throw propper desktop apps at one (I've tried, it wasn't good).
The crop isn't admitting to the ISO it was shot at, but that image of City Hall is showing some pretty nasty pattern read noise (horizontal lines) along with red/green blotchiness.
Sorry, but given that Sigma gains only about a 20% resolution advantage in normal circumstances compared to a Bayer sensor (i.e. you'll get about the same detail out of an 18Mpix Bayer) but the latest generation of Bayer sensors will absolutely kill the Foveon at high ISO, and will give better colour accuracy to boot, there's no way that I'd consider one of these. The idea that Foveon doesn't need an anti-alias filter is a crock also. You don't get colour moire patterns, but you get aliasing artefacts, false detail and all sorts of other horrors.
Again, who said they would last forever? Mechanical parts will last longer IF THEY ARE LARGE AND YOU DON'T USE THEM, but for high duty cycles or miniature parts they will not last as long as solid state by a rather large margin.
Your mech relays required periodic servicing and replacing of contacts. How well would that work on the scale of a DRAM chip?
ALL electronics have a known rate of failure. Mid-life it's so low in solid state devices that you can ignore it, or insure against repair costs for a small sum. The design life of DRAM is beyond the usefull life of a PC so it's unlikely that your insurer would have to pay out.
Now if you want to try building an electromechanical RAM chip to prove me wrong then please feel free, but you'd better have a large warehouse to keep it it and a large support staff to maintain it. Just ask the telco's why they moved to computer based switching, and look at the small boxes that sit in the corner of an otherwise empty old exchange.
Firstly the EU directive only covers faults that were present durring manufacture. If you thrash the SSD and battery mercilessly for 20 months then Apple or any other PC manufacturer don't have to do squat under the EU directive. In fact, under the EU rules you need to prove that it was a manufacturing fault yourself after 6 months. The Apple warranty covers all of that.
Secondly in anyone's book 3 years is longer than 2.
Thirdly your warranty is with whoever you bought the device from (not necessarily Apple), and they are free to nominate whatever repair facility they want. They can also declare it beyond economic repair, and refund the cost LESS an amount to cover the use you've had out of the device. Guess who gets to decide how much that sum is? The retailer.
than your father did, and his knowledge was less than perfect. A mechanical part requires movement to have an effect. As such it is subject to friction, mechanical wear etc. and has a limited number of operational cycles because of this. Old fashioned mechanical relays are rated for up to 1,000,000 cycles for example, which sounds great until you realise that if a modern CPU had transistors that were only that reliable it would fail after 1/10,000th of a second.
A solid state part doesn't need any mechanical movement to work. Thermal effects will have an effect on lifespan (and I didn't claim that the parts last forever) but aren't the only failure mode, and for a mobile class device are quite low. Generally once a solid state device has got past its first few months of life (when it's under warranty anyway) failure rates drop off to very low values and you can expect them to last beyond the useful life of the device it is part of (manufacturers quote reliability in terms of millions of hours MTBF).
SSDs however have basically unlimited read endurance (millions of hours MTBF) but limited write endurance due to the way that pages are erased and then written. They are however socketed so they are replaceable.
You need to check your facts better. Apple still stock 2007 and earlier batteries. At an equivalent age a new battery would take you through to 8 or more years old. How long do you expect to keep using a laptop?
The point I was answering was the claim that dead battery = new laptop, this is patently untrue.
RAM failure? Modern RAM is such dreadfully unreliable stuff isn't it. Why, I've heard that some people can go weeks without it exploding in a shower of sparks. Me, I've not had a RAM failure in the last 10 years. It's a non mechanical part you expect to last well over 5 years.
SSD? Again non mechanical, but with a known ware rate. If you dont thrash the device 27/7 then it shouldn't be an issue though. It's a socketed component (3rd parties will have the connector cloned long before you need one).
Battery? OK, this is the most likely part to need replacing. The cost of a 3rd party, custom formed 95Whr isn't going to be dramatically less than Apples battery replacement charge, plus they recycle the old one.
If you're worried about any of the above then Apple offer to insure repair costs for you for 3 years. You get the first year covered for free.
You do realise that when a DVD drives says that is 16X, thats the PEAK speed. Depending where you are on the disk surface transfer speeds can fall to 1/2 of that.
That aside, the real problem with DVDs for speed is the time they take to spin up and seek to the right piece of data. USB flash is almost instant. With DVD it can take more than 10 seconds to spin up and 0.1 to 0.2 seconds per seek.
Happy feeling? More like the kind of brainwashing that HiFi nuts go through to convince themselves that they can hear the difference between a £5 and a £50 digital interconnect.
I've got a decent separates system with a dedicated HiFi grade DAC (96k 24 bit), and though I can easily spot the difference between 320k OGG files and 256k AAC, I'm damned if I can hear the difference between FLAC and 256k VBR AAC.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
