Re: "What more could you want?"
Erm, if you haven’t read the article then the review is of a Ryzen 4600U based machine with integrated VEGA graphics. Full AMD stack, not an Intel or nVidia part in sight.
2645 publicly visible posts • joined 19 Sep 2007
Given that there’s very little out there that makes use of all that RAM, and software houses are going to write code such that it fits on current mid to high end hardware, then ONLY 10GB isn’t going to be a problem.
The only reason the 2080 had 11GB is because it was a multiple of the memory bus width. The faster GDDR6 in this uses a narrower memory bus, but still gets more throughput because of the newer memory type.
Um, yes. Throwing away 9 rocket engines after only one use is kind of expensive. The Falcon 9 has been tweaked to require the minimum amount of refurb work as well as to land after launch.
The old space shuttle required a huge amount of work between launches and so much was replaced that it wasn’t a practical proposition. The F9 can be turned around in (from memory) 6 weeks and at a low enough cost that SpaceX offer a discount for using a previously flown model. It’s also cheap enough for SpaceX to loft 10 batches of Starlink satellites themselves (so far), while providing cheap ride shares to their partners at the same time.
A quick check shows that the original launch was to an elliptical orbit with apogee of approximately 150,000km and a perigee of 280km. Not that easy to work the numbers out for time to decay.
Modern missions are expected to include a plan to either expend fuel to burn up, or move to a graveyard orbit out of the way when their useful life is over.
It depends on the original orbit. Satellites placed in geostationary orbit are going to be up there basically indefinitely. LEO sats placed around 300km are going to burn up in a couple of years without thrusters to keep them up. That number only gives you a clue as to the initial orbit.
More expensive than mobile BB? Only if you think in terms of 1 or a couple of users for a base station. One base station coupled to a wifi Hotspot could provide for a whole village. Also there are large lumps of the US where even 3G isn’t available, how much have the US companies spent on rolling out their service so far?
The cost of optics is exponentially proportional to their diameter. A space based mirror can be much smaller for the same effective light gathering capacity (no atmosphere to dim and distort the image). As someone pointed out earlier in the thread, the Hubble telescope is only 2.4M, but is still able to look further into space than the largest, most powerful land based devices.
Again, to spot asteroids the main requirement is to cover a lot of space, not maximum magnification. The latest ground based wide-field scope that I’m aware of has a budget of about $28M. It’s not a big step from there to a space based device.
Deploying fibre in rural areas is vastly expensive, and mobile data is still expensive while proving much less bandwidth. Even in the first world (rural USA for example) then sparsely populated areas have little or no (ground based) connectivity. There are large parts of the world where this kind of connectivity would be a godsend.
What they are after is a wide field observation and the comparison of images between different timed shots. Some of the Earth-based versions use standard camera lenses. The mirrors therefore don’t need to be huge, and given a cheap launch it doesn’t need to be vastly more expensive than a ground based model given that they can use smaller mirrors in space.
Erm, Apple’s XCode IDE is also free, and can create code for both the the Mac and iOS platforms. You can write and deploy code to Mac and your own/company iPhone/iPad devices for free. You pay $99/year for individual programmers or $299 for companies if you want to deploy via the App Store. Android is cheaper, but not free (a one-off $25) if you want to use the Play store.
Debit cards are charged as a flat fee. Credit cards charge a percentage of the transaction. From the consumer point of view they aren’t the same though. Purchases by credit card become the CC companies liability if there’s a problem (if, for example, you book a holiday and pay via credit card then the company going bust or there being a similar problem, it becomes the credit card companies problem and they have to refund you), debit cards are like paying cash, and the problem is yours.
The example here was that Epic tried to sneak in a major change as a slipstream update. The idea that minor updates don’t need the same scrutiny is thus disproved.
The other point you seem to have missed is that it’s Apple’s platform, which they are entitled to make a profit from. Epic don’t charge for their games and make their money via in app purchases. If Epic take that in house then where’s Apple’s share comming from?
"You can buy Xbox and Playstation games from many different retailers as you can with Windows and also directfrom teh developers.Yes there are fees for Xbox and PS (games are often $/£10 more than PC."
Erm, you CAN'T publish a game for free for the Xbox or Playstation. You have to agree to Microsoft/ Sony's terms and conditions, pay for the SDKs and development tools, plus pay a royalty for each copy sold. AFTER you have done that then you have to give retailers their own cut to sell it for you. If you want to sell downloads only then you have to sell via the manufacturers own online store.
They are actually making more money out of the deal. On the App Store they get charged $3 on a $10 purchase. If you buy direct then they charge you $8 for the same thing, so unless their transaction cost is more than $0.99 (unlikely, CC companies charge circa 2-3%) they get more in their pocket.
Meanwhile Apple also support purchases by gift card (so they need to give stores a cut), customer refunds (remember the scandal when kids were purchasing using their parents linked credit cards) and curation. It's not all gravy like Epic seem to claim.
Yes, the terrorists choice of champions is ANFO, Ammonia Nitrate (a common fertiliser) mixed with fuel oil, which incidentally was used by the US as the biggest non-nuclear bomb in history for a blast effect test (see https://www.thedrive.com/the-war-zone/35418/this-was-the-largest-conventional-explosion-america-ever-set-off)
4000 series Ryzen mobile CPUs give the 10th gen Intel mobiles a good run for their money in the productivity space (they are a smidgen slower in gaming, but these are not gaming machines), last longer on the same amount of power and are cheaper for a similar config. This Huawei really doesn’t seem to have any advantages over, say, a HP ProBook 445 which can be configured with discrete graphics, 16GB of RAM, 512GB NVMe SSD, Ryzen 7 4750U 8 core CPU and 14” full HD display (Ok, so this is 16:9 rather than 3:2) for $1099, which even gives you change for the $399 Thunderbolt dock with audio conferencing.
That was a heck of a sale, the original list on that lot was north of $4500.
Actually I started with BASIC on a UNIVAC 1110, then MS BASiC on a PET, 6502 Assembler, moved over to Z80 systems, learned Pascal under CPM, added FORTRAN 77 on Boroughs machines, since when I’ve made my way through a stack of languages and currently make a living from C#, SQL and a handful of supporting languages. I’ve no problem with whatever route people came to the field via, I just object to them talking a particular system down without understanding how it compared in context (some machines failed because they sucked compared to the competition. Some succeeded even though they sucked because of software support on them. Some failed even though they were good compared to the competition because the software support was poor).
The Apple II was early to the game, had good software and hardware support, and was competitively priced. For those reasons it deserves some respect.
The H11 had the same 64K memory limit, and at the time 64K was prohibitively expensive. The machines of the day were restricted in their capabilities mainly by what you could afford (and this tended to lead to some very creative programming to get around).
Sure, the IBM (and clones) broke that limit, but it wasn’t around at the time so whatever companies had available to them was necessarily limited.
It wasn’t until the late 80’s that network hardware and hard disks became at all affordable and running an office around PCs became reasonably practical.
The H11 was $1295 for a kit comprising of just the CPU and 4K of RAM. You needed an external terminal and an IO card on top of that as a minimum to make it do anything.
The Apple II was the same price, but all you needed to make it work was a TV and a cassette player. For that reason it sold very well to hobbyists. Following the creation of VisiCalc on the II it’s sales to companies skyrocketed.
The H11 may have had an existing library of software (PDP11 code, on paper tape), but it was much more expensive and took more time and effort to get it to the point it could do useful work, and it didn’t support memory mapped video so was incapable of many of the things that the Apple II became famous for even though it had a more powerful CPU.
Pretty much ANY machine of the period needed to be recapped. The original Mac didn’t sell as well as Apple had hoped, but the Mac line continued to build momentum and the company did rather well from it (until management stuffed things up for them, like the ill fated move into licensing clones.
An engineer is a person who can build for $1 what any damned fool can build for $5.
The genius of Woz is his ability to make something with the minimum number of parts and effort. There were others of the time who were hacking together machines, but Woz worked out how to give the public more for less.
Examples were generating NTSC colour video via deliberate artefacts in a mono picture, using the video system to generate refresh signals to DRAM and designing a floppy disk controller with 1/10th of the chips of then current designs (and that was just for the Apple II).
Ah, the days before a machine needed an active fan even for idling.
Intel were worried about cooling a CPU that drew 15W at full chat. These days a desktop PC will draw anything between about 65W and 135W and will pretty much fall over instantly if the heat sink is dislodged (stopped fans will take them a little longer).
Android outsells iOS in the mobile phone space by around 4:1 then its hard to make a case that Apple have a monopoly in other than iOS systems.
What got MS into trouble was that they had a monopoly in PC operating systems (roughly 90% of the market), and they tried to leverage that to force themselves into other markets (like browsers for example, with the much reviled Internet Explorer versions upto (from memory) 6).
Given that one of their USPs is that they do curate the App Store then it’s not wildly unreasonable that they do wield control over it and enforce restrictions on what may or may not be published.
the Polytechnic which owned the Univac 1110 that my 6th form had time shared access to was struggling with performance, so they bought an upgrade. The engineer turned up expecting to move some jumpers, thus enabling the extra capacity, only to find that someone had already done it. No more performance available.
IIRC they ended up replacing it with a cluster of Prime mini computers.
Most LEO stuff has station keeping thrusters to stop them falling out of orbit due to natural decay (and incidentally give them a certain amount of manoeuvrability to avoid collisions). SpaceX plans to save some fuel/reaction mass to de-orbit their birds at end of life - hopefully Amazon have similar plans.
GPS birds are in a much higher orbit, so are unlikely to be effected.
Internet mostly runs over optical cables, only occasionally over satellite at the moment (the current active birds are geostationary (so at much lower risk anyway), but this causes high latency, so isn’t popular when cabled connection is possible.
TV and Telephone are mostly run over cable or terrestrial radio (although mobile phone base stations tend to use GPS for time synchronisation, so if GPS were effected then they would need modification).
These birds are all in LEO, so should fall from orbit in around 5-10 years.
Any other myths you’d like me to dispel?
Here’s the problem for Intel:
They don’t even come top of the table for SPEC performance. (that honour goes to a Fujitsu SPARC machine tested back in 2017, see https://www.spec.org/cpu2017/results/res2017q4/cpu2017-20171211-01435.html )
AMD are currently making CPUs that in 2 socket configuration need an 8 socket Xeon machine to beat them (see https://www.spec.org/cpu2017/results/res2020q2/cpu2017-20200525-22554.html ).
The Intel boxes are vastly more expensive to buy and to run (all those sockets need lots of power).
There is a limited market for “absolutely the fastest machine you can buy”, mostly companies want the best performance they can afford within their budget, or target a given performance level and then see how cheap they can buy it. Intel have a certain amount of inertia that they can rely on here, as it takes companies a while to test and qualify new hardware, but they are starting to come under fire as the AMD alternatives are looking increasingly attractive. They need to be cheaper and consume less power to compete, but to do so needs a smaller, more advanced process than 14nm, which they haven’t really got (even 10nm isn’t ready for server grade chips yet).
A GeAs chip would have CMOS pairs 3-4 times larger at the same minimum feature size than the equivalent Si chip (because GeAs is not good for hole mobility the PMOS side needs to be 9 times larger than the NMOS, as opposed to 2:1 for Si). Even if the process technology were able to handle GeAs on the same scale as Si, the resulting chips would not have the space for the complex circuits of a modern CPU, and would lack much of the optimisations because of that.
Even if they could reach 60GHz, they would be nowhere near 10 times the speed of current designs, and certainly they wouldn’t be 128 bits wide.
So they are entitled then. It’s only by agreement of the governed over how much.
The problem here is that Ireland is part of a bigger union (just like the US states are part of a union), and they also have to follow EU rules when deciding tax numbers (there is, for example, a minimum rate of VAT that may be charged, and once levied on a good or service it may not be removed).
Yes, Switch Mode PSUs have a small, high frequency transformer in them. Because the source AC voltage is first rectified to DC, then converted to a high frequency square wave, the transformer can be much smaller than the usual 50/60Hz mains transformers of linear PSUs, and you don’t get as much loss from the primary side (mostly due to core eddy currents and resistance of the many turns of wire needed for a 50/60Hz model). Because there is feedback from the output stage to the chopper circuit (that produces the square waves) there is little to no input to this transformer when there is no load on the output.
No, I don’t understand why people are downvoting you either. It seems they don’t understand the (common) technology, but aren’t prepared to put their mouth where their voting finger is by explaining why they think you are wrong.
So why on Earth should they kick a whole department out to an external company, especially when they are increasingly dependent on that department for survival.
In my experience of JL, the problems they have been having are down to the p*ss poor customer support. People are prepared to pay a premium if they know that the support will be good. Our recent dealings have not demonstrated anything like that.
I highly doubt it. Without these you can’t pinpoint your location (and it was the enhanced precision of Galileo clocks that was mostly responsible for cm level accuracy. Block III GPS also has improved clocks for the same reason). You can’t bring these birds back from orbit and retrofit them. Where’s the value in buying a stake in something that doesn’t do what you need?
I’m not a patent lawyer, but things like producing a ringing sound on an incoming call/message doesn’t strike me as new or innovative.
Many of the patent are also old and either expired or due to expire shortly.
Not the US patents system’s finest moment I think.
How about because the 3000 series mobile Ryzen CPUs weren’t very competitive, being beaten fairly noticeably by the Intel range at time of launch. The 4000 series are a lot faster, beating Intel’s refreshed lineup in most test, and using less power into the bargain.
The price that is being asked isn’t that of a bargain basement device, and as such you’d expect reasonable performance compared to similar devices in the price range.
On the basis you are talking about then the NHS is already controlled by a US tech company: Microsoft. Using the technology produced by a company doesn’t mean that it controls you. Apple and Google provide a service to the UK public. The NHS are trying to piggyback off of those services, and must therefore work within the established limits of them. This doesn’t mean that they are controlled by the companies, merely that they have boundaries that they have to work within if they want to add capabilities to their toolkit.
The public in this case has voted with their wallets over what phone they want to carry with them. Do you not think it would be more un democratic to force them to buy some other device because the government can’t do what they want with what the public has purchased?
Really? You do understand that it is based on Bluetooth LE, which will give you a fairly accurate range between transmitter and receiver, plus allows you to dial the TX power down so that the range becomes not much more than the 2 meter limit anyway. There is then a threshold of time for a warning, and another before positive advice to isolate.
I think you don’t understand the technology and how this is expected to work.