Vista's shortcomings are in my face

quot;caching a cache of a cache"

It was a good read, but I think you may have misunderstood how the different caches work.

Disk cache works by loading blocks and not whole files. It doesn't care much for how the file system works, so "notepad.exe" in you example is just a bunch of blocks and the cache will load a bunch more blocks around them (as data tends to be or arranged or grouped together) I suspect the new non-volatile cache on disks will work the same and may even augment the high-speed volatile cache in stead of replacing it.

Memory cache works in pretty much the same way depending on how the layout is structured. Generally related bytes are normally stored together and therefore a high speed cache would load uncalled pages of ram nearby on the premise that the next call will be for those. Again it doesn't care much about the file system in use. I am assuming that this know method will be used instead of something new based on a disk structure.

Oh yeah you forgot about the onchip CPU cache too, so another one to add to your "cash" pile.

Therefore, from what I can see, the only cache that has a direct relationship with actual files is the 'ReadyBoost'. I could be wrong and maybe missing the whole point but I can not actually see any advantage, either my disks are very good and nice and fast or my USB is crap, because I don't see much difference in the speed.

In fact I can only see this being useful when the disk spins down to conserve battery on a laptop and any disk access has to wait for them to spin up.

However the piece isn't about cache, you really just wanted to do the joke.

Leadless

Subject: Asustek confirms distaste for Microsoft Vista

I think they need to sort their licensing out. If we can't transfer vista to our next machine, when we "enthusiasts" change parts more often than we change day's then they are not going to sell it.

I am waiting for a good crack for Vista to allow me to re-activate how every many times i like, so i can wipe, upgrade, or rebuild when i feel like it, or for the Vole to sort out their eula.

Craig

Subject: ps3 gpu blurb

Hello!

As a multiplatform graphics developer I think you are giving the PS3 GPU way too much credit when you call it a GeForce 7900. I'd much rather go with GF 7600, due to the crippled bandwidth in the system, alternatively "a piece of crap". Pick your favourite! ;)

Mag

Subject: 20% more HDDs fail in summer?

I suspect that, rather than power line surges due to air conditioners turning on and off killing the hard drives (why don't the motherboards die as well?), the problem is more likely to be the higher ambient air temperature during the summer, and inadequate airflow over the HDD in the computer case. The HDD overheats (nice fans keep the air moving over the CPU and graphics processor) and it's time for a new disk drive.

Time to back up...

Peter

Subject: The Vistula is wide and deep

I live in Krakow, Poland, near the the Vistula and I can assure you the Vistula is rather narrow (~100m) and shallow (~2m).

Artwi

Subject: NETxEIB OPC Hole

Well, now this is something I know quite a bit about.

The product you referenced is a specific "server" package that was written by a German company NETxAutomation. It basically uses the OPC (OLE for Process Control) standard to provide a vendor-neutral, standardized method for "client" applications, such as a GUI for a process operator to access a proprietary process control device. In this particular case, the proprietary device communicates using the "EIB" network protocol, which is a building automation protocol generally used in Europe, but hardly at all in the U.S.

There are a number of other vendors that provide "OPC" server applications designed to communicate with a wide variety of devices. The current versions of OPC do employ the proprietary Microsoft OLE automation protocols, but the latest version of the standard, which is in the adoption process, will be based on XML and will be decoupled from Microsoft technologies, although most vendors will continue to provide "server" applications that run on Windows (mostly 2K Pro and XP Pro), mainly because that is what most automation professionals current use. Linux is generally not used, because of the need to "roll your own" and most automation professionals do not have the time. We are generally too busy developing control strategies that add value to the bottom line. Pursuing a alternate technology base just because you don't like Microsoft is a luxury in this business.

Thought I would chip in with my 1-1/2 pence worth.

Rich Wargo

Subject: Air conditioners kill hard-drives

It seems the reader who chopped the sh*t out of Nick Farrell, calling the article stupid concerning power issues, was out of line or perhaps considerably uninformed. Why, you may ask? They don't call me Sparks for nothin', baby. Yes, Virginia, I am a full blown IBEW LU#3 Union, A-Division, Journeyman/Forman in THE BIG APPLE, NYC, Hot Cakes. Let me to tell you, some of the installations I've seen/repaired/modded/corrected, etc., would, certainly, more than justify the article.

Because of budget constraints some these poor bastards running these IT rooms are left to run out to the local Staples store to buy something/anything the will give them HALF a chance. Well, UPS ain't enough, tough guy.

Power Conditioning is quite another mater. There are a multitude of ugly beasties that go though the best UPS systems like rats under the rails of the E Train. Harmonics, Sine Wave Distortions, Distorted Phase Angles, Unbalanced Loads, Imbalanced Neutrals, just to name a few. Loosing a neutral in a multi-phase system will fry a UPS in s heartbeat! If you think this stuff can be stopped by your $100, 1000VA UPS, think again.

More to the point, comprehensive power conditioning requires, Inductors, Isolation Transformers, power conditioners, Isolated Separately Derived Grounding, oversized conductors, balanced loads, and finally computer controlled line conditioners. With all this the Engineers still sweat when we hook up the Industrial Grade 3000 pound UPS. This stuff is big, heavy and cost MONEY, big money. Hell, we monitored a huge news organizations power for 6 weeks continuously with state of the art equipment and noise was still sneaking in! Three engineers went over the data for weeks! We discovered that another building's Variable Frequency Drive on THEIR chiller was causing the problem, especially during startup! HELLO!

Nick did a good job of reporting that unseen equipment failures are due to dirty power in lay terms. More importantly, even in the best installations, the root of the problem, which is getting worse, is the power supplied by the utility company's network. Sometimes it's not their fault. Bottom line, it's ugly and it ain't gonna get prettier.

Oh yeah, good job Nick.

Sparks

Subject: Vista compatibility

Just read the Asustek bit.. well, I can second that (well, who can't?).

I am just trying Vista Ultimate for a month and the short comings are in my face, although I am one of those people who only need their machine for 3-4 applications 90% of the time.

So far: I would have to buy a Nero upgrade from 6.x. I can recommend to UK shoppers to pick up anywhich PC magazine this week because they carry the allegedly Vista compatible version of Ashampoo whatever-Burning 6 - the mag costs 3.75.

Quicktime crahes on me all the time, Firefox crashes upon clicking certain links and most importantly, Iomega has not release Automatic Backup Pro for the Rev drive, but only the driver (they don't even mention "Vista" anywhere in conjunction with Backup Pro, so I am slepping another kick in the teeth from Iomega for a good long while in respect to this, are we playing the abandonned customer game again..?).

Did I mention I have to set up an Admin account in oder to play BF2?

Also, it keeps losing the connection for my remote sessions. So even for the little I do this piece of bloat is pretty useless. Thanks MS, can't wait to go back to XP :)

Matt

Subject: AC surges

I see that the great unwashed has made some pronouncements that ignore Ohm's laws. Contrary to Mr Carmitchel's opinion air conditioners cannot cause a voltage spike. They CAN cause a significant voltage drop when starting. I spent the first 6 years of employment working in the field for Houston Lot and Plunder. So I have hands on experience in this area. Motors (Air Conditioning is a set of motors) have inrush or locked rotor currents that flow into the motor to overcome the inertia of being still. This current depending on a number of factors runs between 2.5 and 5 times the normal operating current. The inrush current causes the IR drop to go up by 5 times. For example you may have a 120 volts on your outlets before the start of the air conditioner. When the inrush current hits you may get a surge in amperage which through V=IR causes a voltage drop 5 times what you see after the AC has finished the startup cycle. So for our example let's assume that the voltage stabilizes at 118 v

The advice on UPSs is some of the worst engineering advice I have ever seen on the internet. The commercial UPCs you see down at Best Buy, Circuit City , et al are "buck and boost" UPS's which DO NOT function as a pure UPS. They use relays to switch between the line filter and the battery-inverter combination depending on the voltage sensed on the buss to the outlets. Therefore if the voltage drop is below the threshold relay settings the battery-inverter does not "kick in" to boost the voltage and cut off the line source. Simply notice the number of cycles that you are without power before the UPS kicks in when lightning or other disturbances interrupt your power. It is usually 6-10 cycles so the UPS every one is familiar with WILL NOT provide any protection aqainst voltage fluctuations caused by air conditioners.

EBH