The Apple iPad wifi-only version will be launched on April 3, 2010, according to the official iPad press release distributed this morning by Apple. The 3G/Wifi version will ship ‘in late April’.

Users can pre-order the iPad starting March 13th.

via Gizmodo

Originally appeared on: StationStops - Metro-North Train Schedules and NYC Commuting Blog See the original post here: Apple iPad Official Launch Date April 3, Preorders March 13

The Crucial RealSSD C300 comes in 128GB ($499) aand 256GB ($799) models and has been confirmed by third-party websites as being capable of surpassing 350 MB/s read and 200 MB/s write speeds.

HotHardware.com demonstrated that while the C300 can be connected via the SATA II ports common in most user’s current computers and laptops, it’s performance will be capped. In order to get the full performance of the drive, it is necessary to connect it via the new SATA 3.0 6 Gbps controller.

The good news is that if you have a PCI-E x4 expansion slot, you can add a couple of SATA 3.0 ports and 2 USB 3.0 ports to your existing computer in about 60 seconds with a $30 Asus U3S6 PCI-Express SATA 3 6 Gbps / USB 3.0 combo card (warning: Asus only qualifies this card as working with specific Asus motherboards, but I’ve heard others have gotten it working. I just ordered one and will let you know how it turns out. The thing is, this board is not only less expensive than most competing SATA 3 cards, but the inclusion of USB 3.0 doubles its value, especially if you only have one PCI-E x4 or better slot available).

If you don’t have a free PCI-E slot you will probably need a new motherboard for SATA 3, since the PCI-E bus is the only connector your computer has which is capable of providing the bandwidth required.

Crucial has priced the 256GB version of the drive very competitively at $799, pretty much matching the street price for its nearest direct competition, the
OCZ Vertex 250 GB on a price-per-GB basis.

Pricing for the 128GB model is less competitive at $499, putting the cost per GB at nearly $3.89 (vs about $3.12 for the 256 GB version – someone check my math)

As of this writing, the usual suspects – Amazon and NewEgg, did not have a listing for the drive. It will be interesting to see if they discount it at all over the Crucial store.

Originally appeared on: StationStops - Metro-North Train Schedules and NYC Commuting Blog See the original post here: Crucial C300 SATA 3 6 Gbps SSD Now Available On Crucial.com

Unfortunately, some of these technologies are going to be unavailable to most Mac users for some time. Its one of the disadvantages of Mac ‘s hardware being tightly controlled and locked down by one company, and the relatively slower approach Apple takes when deciding which new technologies are important and when.

Let’s take a look at some of these technologies, why they matter, and how Mac users – especially buyers of Apple’s latest iMacs – are largely locked out of them due to the lack of desktop expansion options common with PCs. In fact, several new models of PC motherboards are already in the channel which, when combined with Windows 7, include complete support for all of them on-board.

USB 3.0

The most important is USB 3.0 To date, modern hard drives have been severely bottlenecked by Firewire and USB 2.0 as convenient external drive connections. eSATA has been a help, but as many users know, eSATA (if Apple even decided to include such ports) can be fussy and its usually easier to just use USB 2.0.

USB 3.0 is about 10x faster than USB 2.0 in the real world, which really lifts the bottleneck off of common external storage. This is a big help especially for Time Machine users who allow Time Machine to back up their Mac frequently.

But there is another great benefit to USB 3.0 also – and that is the extra electrical power it supplies to devices, which makes it such a better choice for external storage over all other methods.

With USB 3.0, it is possible to connect much more power-hungry external drives without a separate power connector or y-cable. As every computer user knows, the fewer cables the better no matter whether where you are, but mobile users will especially appreciate this.

Now, if your notebook – Mac or otherwise – has an ExpressCard port, you will likely be able to add an awkward adapter to enjoy USB 3.0 down the road – but you will need an external AC adapter to supply the extra power it needs which it cannot draw from the ExpressCard slot.

The problem is, there are a lot of very popular MacBooks without an ExpressCard slot out there (mine is one – a 2008 Unibody MacBook) – and iMac users are sadly completely out of luck, as that machine has no PCI Express expansion.

SATA 3.0 6Gbps

SATA 3.0 6Gbps is less important to the average user, but will be a nice feature for power users very soon. The reason is that this month the new Micron C300 SSD will be released which will be the first – but definitely not the last – to exceed the real-world throughput of SATA II.

Since this performance boost is largely delivered by the drive controller, and not the quality of the Flash memory, it is not unlikely we will see more drives at reasonable (relatively, for SSD) prices over the next 12 months.

Again, if you have an ExpressCard slot you may be able to take advantage of this for external storage – but ironically, not for internal storage. In general, due to their price, most users (like myself) prefer to use SSDs as internal boot drives and traditional drives for external bulk storage.

And while we are on the topic of SSD’s…

High Performance Solid State Disks (SSD’s)

MacBook users are not completely locked out of the great performance advantages of the newer High-Performance Solid State Disks. I am writing this on a MacBook with an OCZ 120 GB Vertex SSD right now, its been in their for 9 months with absolutely tremendous performance and zero issues.

However, unlike Windows 7, MacOS is not ‘aware’ of SSDs and treat them as a normal hard drive. This has some drawbacks.

Most important is the lack of TRIM command support in MacOS, which is supported in Windows 7. The TRIM command enables the OS and SSD to negotiate the management of drive space to avoid specific performance issues with SSDs when they start to become full.

Whats important to remember is that the performance hit from not having TRIM support will vary from drive to drive, and will reach a certain point after which it wont get any worse. I have read few anecdotes from people who have observed such a hit that they sold their SSD. I personally have never really observed it on my MacBook, but I have never come close to filling the drive either.

My Windows 7 SSD *is* nearly full, and I haven’t perceived any issue with that drives performance either.

(This is potentially relevant however since most SSD’s are 120 GB or less and are very expensive. As a result, users tend to buy as little storage as they can tolerate, and as a result will typically fill it faster than a $75 1TB Hard Drive.)

Windows 7 will also make other adjustments in how it manages an SSD which involve the disabling of features which were only designed to overcome the performance profiles of a traditional hard disk which are not present on an SSD.

For example, an SSD does not need to be defragmented (and shouldn’t be since you want to avoid unnecessary writes). Other features such as Superfetch, boot and application launch prefetching, ReadyBoost and ReadyDrive – which are designed to overcome the access time variance with hard drives – are also unnecessary.

These steps are also important to the lifetime of the SSD, since fewer writes = longer SSD lifetime. (In the real world, however, SSD cell write wear is not going to be a problem for most users – and modern SSD controllers can reduce it’s likelihood enormously with wear leveling and write amplification reduction algorithms)

Note: Intel claims that the average user writes about 2-3 GB of data to their hard drive every day, and that in order to wear out an Intel SSD, you would need to write about 100 GB day for the drive to wear out over 5 years. Extrapolating, this means it is more likely you will die before you drive suffers a write-wear failure. So, although SSD cells do in fact have a limited write capacity, its relevance is largely mythical. Compared to the reliability of a mechanical drive, its nearly hysterical.

Finally, Windows 7 will avoid the unnecessary use of unused disk space, properly align the NTFS partitions, eliminate merge operations and prioritize garbage collection (PDF of relevant PPT Presentation).

So, Windows 7 users are going to get better performance and better reliability out of SSDs compared to other OS’s. It really is entirely up to Apple to start taking similar steps.

iMac and Mac Mini users have a special problem with SSDs however, and that is that neither allows the user to (easily) replace the hard drive themselves. I certainly would not recommend any user not familiar with building their own computer crack open their Mac Mini with a Spackle knife, or attempt the challenging and delicate disassembly and thermal sensor handling required to simply change the hard drive in a 27″ iMac.

Also, there is only one drive bay in most Macs, so the preferred desktop strategy of having an SSD boot drive and an HDD mass storage drive onboard is impossible (another reason the iMac would benefit from USB 3.0 – the HDD could be external without the performance penalty).

In Conclusion

Mac users seeking to upgrade should be aware of these new technologies and decide for themselves which are important to them over the lifetime of their next Mac purchase.

Unfortunately, unlike other computer manufacturers and OS’s, Apple keeps its roadmaps under its hat, making an upgrade decision frustratingly difficult compared to PC users. Which makes it the same old game of wait-and-see.

Originally appeared on: StationStops - Metro-North Train Schedules and NYC Commuting Blog See the original post here: Apple Mac Users Locked Out of USB 3.0, SATA 3.0 6Gbps, and SSD Support

I have always used solid color backgrounds for my Windows desktop, usually something really dark, like a very dark green or blue or even black, to minimize distraction.

It’s also a habit left over from the days when a fancy desktop background consumed precious memory and boot time from Windows.

Well, for some bizarro reason, Windows 7 (and Windows Server 2008) has this bug where if you use a solid color background, it stops and twiddles its fingers for a fixed 30 seconds at boot time.

When I read about it, I immediately benchmarked my Windows 7 box, and sure enough, my boot time was 50 seconds with my solid color background, and 20 seconds with Windows’ high-res wallpaper of a giant bale of hay.

Which answers a lot of questions I had about why it was taking so long for a quad-core computer with a high-performance solid-state drive to boot.

Obviously, this problem is easy to fix by using wallpaper – even if its an image of a solid color.

There is also a Microsoft hotfix – but I am not so big on installing a hotfix for something with such an easy workaround, and would rather wait for a fully-qualified update to fix it…

Originally appeared on: StationStops - Metro-North Train Schedules and NYC Commuting Blog See the original post here: Windows 7 Solid Color Background = 30 Second Boot Delay (!!)

The worst part of Windows for me is *activation*.

Activation is the secret mojo in Windows which gets all pissy if you change hardware in your computer too often and makes you disgrace yourself by literally making a phone call to Microsoft headquarters and asking their permission to keep using it.

First, you have to read off a painfully long sequence of numbers and letters to an automated voice system. Sometimes the automated lady is OK with the numbers you read off, and sometimes she’s not. If she’s not, then your’e going to get transferred to an call center and read some more numbers ALL OVER AGAIN to a tech. Chances are you will have to do this multiple times as you will be stepping on his words because of the phone delay to India (he’s trained how to avoid this, you’re not).

The third level of hell comes if this guy doesn’t like your numbers. At this point, statistically speaking, you are probably running pirated software. But there are tons of people who aren’t. There just isn’t anyway to be sure.

In my experience, he’s going to ask some questions, but depending on the version of Windows license you have (OEM/Vendor/Retail), how dramatically your hardware changed (reflected in the number sequence), and how often activation has failed before on that product key, he’s going to go ahead and give you a pass.

Which requires – what else – that he gives you another really long sequence of numbers and letters to type into the machine to allow you to use the software you paid for, now that you’ve lost an entire morning in productivity at a cost higher than the price you originally paid for the software.

This week I was swapping some drives around and I got hit with the activation message again. Luckily the automated lady was OK with my numbers and gave me a pass this time. But the fact that I even have to do that for any software is just really infuriating.

Originally appeared on: StationStops - Metro-North Train Schedules and NYC Commuting Blog See the original post here: Windows 7 Didn’t Fix The Worst Part of Windows

I was new to the Valley, and a bit bright-eyed (that would fade FAST). For a new transplant, I was looking forward to visiting the holy Cisco campus and hanging out with other developers of the first generation of internet superstartups to discuss Linux development with its inventor, Linus Torvalds.

He might as well have been God.

But I learned quickly that Linus is very open, honest, and frank about Linux’s shortcomings – much more than most of his followers.

He went on to talk at length, in fact, about the superiority of the Windows CPU scheduler, and how Linux had not yet been able to touch its performance. It was very important to him that it improve, but confessed that it was one of the most difficult problems to solve technologically.

[The CPU scheduler is the part of the operating system kernel whose purpose is to efficiently assign slices of time to all the running processes to the CPU(s)]

This was kind of disconcerting to some of the attendees, as there has always been this aura of technical superiority of the Linux kernel to Windows. But the reality is, and remains to this day, that all of the major operating systems have their advantages and disadvantages, and to religiously ignore that fact is to just ignore reality.

Recently, there was another big problem uncovered in the Linux kernel scheduler that remains unfixed in all major distribution releases today. Of course, as always there is the option for the user to download or compile newer forks of the kernel which fix the problem, but that is not an expectation of everyday users, nor is it something desirable to IT departments for whom stability and uniformity of production OS software is essential.

Users had started to notice that the popular x264 video encoding software, which is one of the most popular, if not the most popular applications which can truly exploit efficient CPU scheduling, just ran a lot faster on Windows then Linux on the same computer.

Now, we all like software that runs faster, but video encoding is a special case of where performance is far more important than most because of the time required. You might not even notice if your browser rendered a web page 50% faster between one page load and another these days, but that same level of improvement is a godsend for video content creators.

As it turns out, changes in the Linux scheduler over time had introduced performance problems for some multithreaded applications. For most of these apps – like a webserver for instance, the issue impact is negligible, so the problem was unknown for some time. But for x264, the problem was enormous.

How enormous?

Well, the version of Linux kernel most users are running today are likely experiencing a performance penalty with x264 of up to 70%. For a rendering farm this is a massive cost penalty, as they would logically need to spend 70% more on hardware to meet the exact same workload requirements.

The problem was so severe and obvious that once notified, Linux kernel developers checked in the fix overnight, and a patch was issued. Since there attention was already on it, kernel developers were to able to later add even more performance to the application through additional kernel changes.

However, Linux distributions like Ubuntu do not run the version of the Linux kernel which was built last night, they run a proven and stable kernel which has been around for a while, with good reason. Enterprise Linux versions from Red Hat and Suse tend to use these stable kernel versions even longer.

So, for one client, who does about 18 CPU-hours of x264 encoding on Linux DAILY, this patch was really important, as it returned large gains of efficiency instantly with no new hardware costs. But lacking a vigilance of the monitoring of kernel patch developments, the problem and solution had existed outside of our awareness.

Originally appeared on: StationStops - Metro-North Train Schedules and NYC Commuting Blog See the original post here: Linux Kernel Issues with CPU Scheduling and x264

Recently I have been looking at ways to reduce energy usage around the house.

One long overdue way was to put my always-on loaded 300w PC Desktop to sleep and migrate its server duties to an always-on Mac Mini, which probably idles around 16w.

Despite all the cool stuff it does, the mini is still stock, 1.86 GHZ with 1GB of RAM. It could probably use a little more RAM for running Boxee, but besides that its fine.

I also located this Mini at my television, so I could use it as a media server as well. This is where a Mac Mini is a better solution than a cheaper, dedicated NAS server. Here are just some notes on what I use the Mini server for, and some gotchas to look out for when staging your own home mini server.

1. A Time Machine server.

My MacBook will backup to the Mini over the network whenever its plugged in, or whenever I tell it to. The Mini itself also backs up via Time Machine to its external hard drive.

2. A PC backup server.

Infuriatingly, Windows 7 has a great backup program built-in, but it will not allow you to use a network drive with a Home version of the OS. Like remote desktop, this is an essential OS service which Microsoft insists on using as a carrot for home users to pay twice as much for Windows 7 Professional.

So, instead, I use a third-party backup software.

3. An iTunes Server.

I could spend a few paragraphs bitching about how difficult iTunes makes managing music for users with multiple computers and devices, but I suspect most users know this already.

Still, its a big plus to have one ‘master’ always-available iTunes library, and drag and drop stuff from it onto my laptop whenever I want to put it on my iPhone, rather than keeping all the data on my laptop.

4. A Boxee server.

I have a love/hate relationship with Boxee. My #1 complaint is that while it does so much, navigation and stability is poor.

The most annoying problem with Boxee is that you cannot customize your home screen. If I could just put shortcuts to my favorite Pandora stations and Hulu Feeds on my home screen, I could cut my navigation on Boxee by 90%.

Which is important, because Boxee navigation is kind of a pain in the butt with a 10-ft setup. There are lots of ways to do it but most of them aren’t that great. Plus, on a 2008 Mini Boxee can be sluggish.

You can use the Boxee remote iPhone app, Mobile Airmouse iPhone app, or a Mac remote to control Boxee. There is a Boxee widget also, which I have installed but not tried yet – I like the idea though.

The problem with using iPhone apps for control is that you need to navigate to and open the app on your iPhone whenever you need to do anything.

The problem with the Mac Remote is that I find it kind of unreliable and it will also control my notebook in the same room. Then there is the conflict with Front Row, which I also use.

Finally, I often check into my Mac Mini to find that Boxee has crashed with a dialog box on the desktop.

On the plus side, Boxee has some nifty features – its Pandora client is great, and I really like the ability to pull up MTV videos when browsing your own music. Hulu integration is klunky but it works.

Also, forget VNC-ing to a box running Boxee in 1080p, performance is horrible.

5. Streaming Media Server

Using Simplify Media, I can stream my iTunes library to my iPhone over the internet. Using StreamToMe, I can stream video as well. Both have some limitations but its great to have.

Simplify Media’s main problem is its constant indexing delays. On their website, they seem to kind of punt this issue off, but its the #1 complaint in the reviews. When they launched their version 2, they really needed this to go away.

There are some pluses to Simplify Media though, like sharing your library with friends and shuffling songs, as well as not requiring port forwarding for internet use.

StreamToMe is a little more straightforward, less feature-packed, less annoying, but also streams video, which is a big plus.

6. General-purpose Network-Attached Storage (NAS)

I like to run SSD’s on my workstations, but obviously they don’t have much storage. with network storage, you don’t really need much local storage.

7. Web/LAMP Server.

I use my MacBook for web development, but am moving this to the Mac Mini.

8. A VNC server. Mac OS comes with a VNC server (aka ‘Screen Sharing‘), but its just non-standard just enough to make a swap out worthwhile. Two of the major issues with the Mac VNC server is that it is not automatically compatible with general-purpose VNC clients. Another is that it is has relatively mediocre performance.

By turning it off and replacing it with OSX VNC-Server (aka Vine Server), you get a faster, more compatible remote access solution. On the other side, try running JollysFastVNC as the client for even more performance (on the PC, I prefer UltraVNC as client).

The downside of this is that Finder will not automatically give you the screen sharing option for a computer running OSX VNC-Server, and I doubt “Back to My Mac” would work. You can get around this by leaving screen sharing as-is and setting up OSX VNC-Server on port 5901.

My main gripe with VNC for remote access in general it that you have to run it at the same resolution as the main display, which is rarely what you want.

For example, since my Mac Mini is connected to an HDTV, its display resolution is 1920×1080. This is a problem when VNC-ing in from my 13″ MacBook, as the display needs to be scaled and is hard to read. It also lowers overall VNC performance.

JollysFastVNC has a nifty feature where you can run a kind of magnifying glass over the parts of the screen you want to use, buts its still awkward.

This is where I really prefer Windows Remote Desktop. When you connect, you can run it at a resolution which makes sense for the client, and it has much better performance – much better than the best VNC setups.

One plus of VNC over Windows Remote Desktop is that it doesnt boot the display to the login screen while you’re connected remotely.

Hooking It All Up

If you are going to hook your Mac Mini up to an HDTV, you are probably going to need some special cables you don’t already own for the best results.

Video Hookup

The Mac Mini is capable of 1080p, but it has a DVI-out, which most HDTVs do not have an input for (but absolutely should).

You can use the DVI-to-VGA converter included with the Mac Mini for connecting to an HDTV’s PC-in port, but for some reason HDTVs can be really picky about what resolutions it lets a VGA-connected device run in. In the case of my Panasonic Viera G10, max resolution is 1400-something. Its usable, but native 1080p just looks so much better.

The answer is a DVI to HDMI adapter. , which works perfectly. In my experience, the Mac Mini will just automatically switch to 1080p with this cable, and looks outstanding.

The downside of connecting DVI to HDMI is that your TV will expect the audio to come through the HDMI port as well. It is very rare for a TV to accommodate an HDMI connector and allow you to associate it with an alternate digital audio source, which is really annoying.

I get around this by using a separate audio input on my receiver, but this is also inconvenient as all of my other TV inputs are completely switchable from the television remote.

Audio Hookup

The Mac Mini and Macbooks have special mini-plugs for digital audio out. This is confusing, as consumers are used to using TOSLink or RCA PCM connectors.

Of course, Apple does not include the special cable to do this with the Mac Mini. I can’t even find it on their website. What you want is TOSLink to MiniPlug adapter (this is the one I use). Like a lot of adapting cables, I have no idea how this magic works.

The Mac Mini will not boot without a keyboard attached, which is an annoying gotcha for a server. I just plugged in an old USB keyboard and threw it behind the entertainment center. The mini will still complain about the mouse, but it will boot. You can either plug in a mouse or just ignore it like I do.

External Storage

As for external storage, I just relocated my existing 1TB disk drive to a cheap external firewire enclosure. You will hear a lot of argument over Firewire 400 vs USB 2, but truth be told I have noticed very little if any difference.

I formatted the disk as Mac OS Journaled to avoid issues. The downside of this is that I cannot connect the external to a PC and copy the contents if I needed to do that. FAT32 will give you the file size limitations you don’t want on a backup server. From what I understand, you can modify your fstab under Snow Leopard to mount an NTFS volume as fully read-write, but having it be a native Mac OS volume hasnt really created any issues for me. I also suspect Time Machine would have an issue with a non-Mac partition.

Network Connectivity

Network connectivity is very important for a server – it moves a lot of big files. I use a Netgear Moca adapter to connect my entertainment center to the network. Ideally, I would get it connected to GigE but I dont have a way to pull Ethernet to my living room.

Since most of my remote connection to the server will be via wireless or internet, it doesn’t really make much difference.

Network Routing and DNS

For a lot of reasons, I prefer to use static DHCP IP assignments for all the devices on my network, and to give them local names assigned by the router.

Most people are used to setting up devices with DHCP, which basically just asks the router for an IP. With Static IP addressing, you can have the device request a specific IP. This provides a consistent IP assignment, which is helpful.

I prefer to use Static DHCP, a third option, as it solves a lot of problems with the first two. With Static DHCP, you specify on your router which MAC addresses get what IP address. Then you can assign a name to it in the router’s DNS server. For your devices, they think they are just using regular DHCP, which is probably how they came shipped, and will work predictably when you take them outside your network.

For example, my Mac Mini’s Ethernet port is setup for DHCP, but my router’s DHCP is setup to always assign the same IP address to that Ethernet port’s mac address. The DNS server on the router is then setup to assign the name ‘macmini-ethernet’ to that IP address.

Now I can connect to the macmini from any device on the network as ‘macmini-ethernet’ without modifying its local HOSTS file.

When I login to my router, I don’t have to guess which IP/MAC addresses shown as connected on my network are – they all have names assigned by the router independent of whatever that device thinks its name is.

You can also now edit port forwarding settings by device name to insure that if your device changes IP for any reason, the port forwarding still works.

Originally appeared on: StationStops - Metro-North Train Schedules and NYC Commuting Blog See the original post here: Notes on Running a Mac Mini as a Home Media Server