Thursday, May 14, 2015

Scientific Linux 7.1 review






Since my last review of Scientific Linux 7.0 I've added and additional 8 GB of RAM and the 250 GB SSD, my buddy has likewise upgraded his system with an additional 8 GB of RAM and added a 250 GB SSD as well. Both machines are mirror images of each other built with EXACTLY the same parts. If something goes wrong we can test it out and find out if it is USER ERROR or something more basic.

Let's start with the GOOD news: Scientific Linux 7.1 is vastly improved over the disaster known as Scientific Linux 7.0. The coders and developers at both Red Hat and Scientific Linux need to be highly commended for their efforts they put into 7.1. The BAD news is Scientific Linux 7.1 is UNPOLISHED and still needs a lot of work.

One of the advantages of having a pair of mirror image machines is you can spot problems that would normally have you tearing out your hair trying to solve. I had no problem this time around doing the initial install of Scientific Linux 7.1. It was only after I did the install that I started having troubles. The single biggest problem was in adding the latest version of kmod-nvidia drivers. I also ran into issues installing VMware Player 7.1.0 -- which installed just fine... but then crashed the first time I started it up. It was the problem with the Nvidia drivers that would throw me. I found every hack in the book, but no matter what I did, the second I rebooted the machine I got a Black Screen with a blinking cursor. After 5 complete installations I was set to throw in the towel and live without the Nvidia working.

While I did my install using manual partitioning, my buddy did his using LVN partitioning, but did not want to lose his Windows partitions. He got the same results as I did. So we decided he should try to manually partition his SSD. He never made it past the point where he could install the OS, which I breezed through. We decided to check into BIOS. The first thing we noticed is that he had two CD-ROM drives listed, not one, even though he only had one physical device: One UEFI and one NON-UEFI. The UEFI was listed first, then came the NON-UEFI. Just for fun we had him change the order and placed the NON-UEFI CD-ROM first. He then rebooted and had ZERO problems with the install. Next he then found an interesting link in regards to installing kmod-nvidia drivers in Scientific Linux 7.1, and followed the instructions, and PRESTO he had Nvidia up and running.

http://linuxsysconfig.com/2014/09/nvidia-drivers-on-Scientific Linux-7/

Excited by the news, I jumped on it, only to end up with a Black screen and a blinking cursor. One other difference in our install procedures was I installed the whole thing -- 7 GB -- which required that I install it from a thumb drive, while he installed his from a DVD live version. In BIOS the thumb drive shows up as a UEFI hard drive. Based on my buddy's experience I went over to the "Boot" portion of the BIOS and found in there a NON-UEFI version of the thumb drive. While I could not put this version in train I could over-ride the boot sequence and boot directly from this NON-UEFI version. The first thing I did was to install the LVN version and followed the hack my buddy found... and like him my system came up with Nvidia. I then went back and installed a test version via manual partitioning, but first I checked out to see what partitions were created by the automatic LVN partitioning scheme. What was created a a /boot, a /home, a /, and /swap; where as in all my installs there was also a /boot/efi partition. This time I left out the /boot/efi partition and the software installed just fine!!! Also the hack worked just fine and and the kmod-nvidia drives installed just fine as well. For the 8th time I did a total re-install only this time I booted from the NON-UEFI drive and left out the /boot/efi partition during the install.. It is just a guess but I suspect that Microsoft with its undue influence over everything, everything now defaults to UEFI, to include UEFI devices, motherboards, etc, and these receive priority over non-UEFI items, thus if you have a recent vintage motherboard, or other devices, they are set up to comply with the Microsoft UEFI policy, but to be backwards compatible, there is a list of non-UEFI items. If you are installing Scientific Linux 7.1 your first trip should be to BIOS, to ensure that whatever device you are booting from is NOT UEFI ENABLED. If you are doing Custom Manual Partitioning DO NOT install a /boot/efi partition -- you do however need a /boot partition. Once you have done that the rest of the install is rather straight forward.

The BIOS hack needs to be published and save everyone from tearing out their hair in frustration. Were it not for the fact I was working with a mirrored computer I might not have caught this.

The biggest problem with Scientific Linux 7.1 is that it is grossly unpolished. There is no reason that you should have to jump through a bunch of hoops just to install kmod-nvidia. A "yum install kmod-nvidia" should be all that is needed. Other problems come to the installation of Users and Groups. In Fedora, all you do is go to Administration --> click on "Users and Groups" and be on your merry way. Both in Scientific Linux 7.0 1406 and 7.1 1503 there is no easy way to add Users and Groups, as the administrative icon is missing. If you want to add Users and Groups you need to be Old School and do it from the CLI. That said I was able to add the Users and Groups icon back in: You need to 1) enable several repos including elreo, epel, extras, and nux-dextop. 2) Run yum grouplist 3) Run yum groupinstall "MATE Desktop", yum groupinstall "Xfce", yum groupinstall "Milkymist", and groupinstall "Haskell". It is somewhere in one of those four groups. When you next click on Applications --> Administration you will find the familiar "Users and Groups" icon.

What about tunes?!? If you are looking for kscd, forget it, I've looked and still can not find it. The news is slightly better if you are a fan of Amarok. IF you plan to run amarok do try to install it before you add your kmod-nvidia drivers. Before I "fixed" the OS to run kmod-nvidia, I had no problem installing amarok, and it worked just fine; after I "fixed" it so I could run kmod-nvidia I went to install amarok I got the following message:

Error: Multilib version problems found. This often means that the root
cause is something else and multilib version checking is just
pointing out that there is a problem. Eg.:

1. You have an upgrade for qtwebkit which is missing some
dependency that another package requires. Yum is trying to
solve this by installing an older version of qtwebkit of the
different architecture. If you exclude the bad architecture
yum will tell you what the root cause is (which package
requires what). You can try redoing the upgrade with
--exclude qtwebkit.otherarch ... this should give you an error
message showing the root cause of the problem.

2. You have multiple architectures of qtwebkit installed, but
yum can only see an upgrade for one of those architectures.
If you don't want/need both architectures anymore then you
can remove the one with the missing update and everything
will work.

3. You have duplicate versions of qtwebkit installed already.
You can use "yum check" to get yum show these errors.

...you can also use --setopt=protected_multilib=false to remove
this checking, however this is almost never the correct thing to
do as something else is very likely to go wrong (often causing
much more problems).

Protected multilib versions: qtwebkit-2.3.3-3.el7.x86_64 != qtwebkit-2.3.4-3.el7.i686

I suspect the problem occurs because when you run the hack that "fixes" it so you can install kmod-nvidia part of it requires you to install 32 bit libraries. To quote the source:

"When running a 64bit OS, the 32bit Nvidia libraries may also be needed for compatibility, I always install them. The good thing is that kmod-nvidia also disables nouveau automatically, so no more manually tweaking modprobe and grub

"yum install kmod-nvidia nvidia-x11-drv-32bit"

I suspect that the "!= qtwebkit-2.3.4-3.el7.i686" is 32 bit and that conflicts the amarok install. You might try to get around the problem by installing amarok before you fix the nvidia drivers problem. No guarantee, but it is worth a shot.

The alternatives included in Scientific Linux 7.1 are sparce -- Rythembox being the ONLY alternative.

Another place where Scientific Linux 7.1 falls down is when it comes to turning ON and turning OFF services. You use to go to Applications --> Administration and find "Services", you go there open it up and then at a glance you could see what was running, and what was not. You could also choose to STOP and/or RE-START any given Service. Like "Users and Groups" this too is missing. There may or may NOT be something similar in Scientific Linux 7.1: If you go to Favorites --> System Settings --> find System Administration --> find and open Startup and Shutdown --> Service Manager you have something that looks sort of like the old service manager. things such as ntp, etc, are missing. The *old* version was far better, and for better or worse, it should be included as an alternate to the "new" version.

One place that is a clear improvement is VMware Player. Once I "fixed" the problem with installing the kmod-nvidia drivers, it also it seemed to "fix" the problem with the Vmware Player. It now opens all my Virtual Machines without a problem. One BIG improvement was that even though during the install the 1 TB SATA is not in the train (it is listed as being found, just not listed in the boot sequence)Scientific Linux 7.1 goes out and can see all the directories on that HD. You just need to select the directory, enter your secret sauce root password, and BINGO!! you now have access to the data on your other drive!!!

OTOH if you are looking for Games -- card, or board -- SORRY you are out of luck. I've gone in search of these and none can be found.

CONCLUSION: Unlike Scientific Linux 7.0 1406 which was a Total Disaster, Scientific Linux 7.1 1503 is very much a usable OS though still lacking in features. It is clearly NOT designed for a newbie who would through up their hands in total defeat. It really is a DIY type of OS -- if you are willing to sink a lot of time into it, it really is a nice OS with some nice features, but there are clearly places where it can be improved: It still needs a way for people to Add Users and Groups; kmod-nvidia should be a rather straight forward install without needing to jumps through hoops; amarok, kscd, should be part of the multimedia package set as these are two old standards people know and love; games -- totally missing and found on every distro except Scientific Linux 7.1 (and probably RHEL 7.1) need to be added; Administrative tools such as the "classic" "Services" that was found under Application --> Administration should be reinstalled; and overall the OS needs to be polished. In short Scientific Linux 7.1 needed far too much hacking to be made usable. Unless I encounter other problems I can now say I will be migrating off Fedora 20 and to Scientific Linux 7.1. While it still needs work -- a lot of work -- Scientific Linux 7.1 is quite usable, even if not as polished as Fedora 20, and 21. The Problem with Fedora, is you are always on the Bleeding Edge -- and trust me you do a LOT of BLEEDING; Scientific Linux by contrast is nice and stable for the most part, however Scientific Linux 7.x has been quite a disappointment. With the release of Scientific Linux 7.1 1503 I can now say it is usable even if it still needs work. For the Newbies out there I would pass on Scientific Linux 7.1, and wait until Scientific Linux 7.2 is released when hopefully many of the problems I have encountered have been fixed; for the more experienced user with a handful of tricks published above I think you'll really enjoy this release, understand this is closer to a Beta or a RC1 release than a final release since it still needs work.

Monday, May 11, 2015

Nintendo Amiibo isn't worth purchasing

 Amiibo and Club Nintendo aren't worth it.  Club Nintendo didn't have good enough free Virtual Console games and the known accessories are complete junk.

We’re right in the middle of a storm. Nintendo botched it big time when it came to stocking the new Amiibo figures with any sort of sense, with such ridiculous, half-assed explanations it isn’t unreasonable to assume the supplies are scarce on purpose. The moment folks got wind of certain Amiibo figures being hard to get, the scalpers and desperate consumers declared war and dumb things are happening as a result. People are now importing Amiibo, re-selling them for several times the retail price and of course paying several times the retail price. Other people are just unhappy and/or disappointed. Add in children, a huge part of the Nintendo fanbase to the equation, and things just get sad.

When you get caught up in a craze like this, it’s easy to forget what’s important. Like, for example, if the damn things are even worth buying. Maybe you’re a fan of the whole NFC thing as a genre, and know exactly what you’re getting into. Maybe you couldn’t care less about Skylanders or Disney Infinity, but the adorable Nintendo characters are shooting daggers into your heart. Either way, buying into this stuff is a hell of an investment depending on your approach, and it’s absolutely worth discussing how you’re going to justify drop-kicking somebody’s mom to get a hold of Best Buy’s last Little Mac. Here are a few talking points.
Price

Price is definitely a factor for anything, but the scarcity and instability of the secondhand market right now is making these things fluctuate everywhere but the unreliable retail locations. If you can find what you want at Toys R Us though, you can do pretty well. At least for the holidays, you get a slight discount when you buy two Amiibo figures. Normally, they’re $12.99 each. That’s pretty good in terms of NFC figures; Disney Infinity toys are a few dollars more. If you want to import to get the more hard to find Amiibo, you’re probably looking at around 20 bucks and shipping. The secondhand market is nutty, going as far as 50 dollars or more for the rare figures, like Marth. Do not pay more than around 20 dollars for an Amiibo. Do not pay more than around 20 dollars for an Amiibo. You will be disappointed.
Quality

Unfortunately, Amiibo seem prone to factory defects and paint issues. This makes pre-ordering thoroughly unappealing, which is awful considering how that’s going to be the only option for a lot of these things, unless they get restocked later. It’s still vague as to whether or not that’s going to happen, and to what capacity. Thanks, Nintendo! That said, when you get one without any problems, they’re pretty cool. Don’t expect the same level of artistry as the Disney Infinity figures, however. Those things are amazing. Amiibo are still nice pieces, and some of them have a lot of neat little details, like Peach. Some of the less complicated characters, such as Little Mac, are going to be a little boring if you’re not into the character or franchise. It’s also worth nothing the stands on some of these figures. Some of them are a hideous shade of yellow for no clear reason, and others just have them in visually awkward places. Mileage may vary on characters like Link and Captain Falcon.
 Function

Here’s the big one. In terms of actual, NFC video games thingy function, Amiibo are… kinda garbage. Especially compared to the competition. Right now, three games are Amiibo compatible: Mario Kart 8, Hyrule Warriors and of course, Super Smash Bros. for Wii U. In the first two, Amiibo simply unlock some in-game items, mostly cosmetic. Using the Link Amiibo in Hyrule Warriors does net you a new weapon, but the others just give you a random item. In Mario Kart 8, you get outfits for the Mii racer depending on which characters you use (not all of them are compatible). The functionality in Smash Bros. is a bit more elaborate, with the Amiibo becoming trainable A.I. which you can customize and fight with.

What’s going to make or break Amiibo is how Nintendo supports them after Smash Bros. When you look at Skylanders or Disney Infinity, you can immediately see the value. Buying a figure gives you tangible in-game content. If you pick up Hulk, you can play as Hulk. When you bust your ass to find a Little Mac, you could already play as him in Smash Bros. Unfortunately, with all the artificial scarcity going on, it’s hard to imagine a game revolving around the figures working out very well. If Amiibo continue to unlock alternate costumes and items, that doesn’t sit well with me either.
What does “Amiibo” even mean, anyway?

Ultimately, I’d be willing to argue that without the supply issues, Amiibo probably wouldn’t be the hot topic it is now. Does that mean Nintendo didn’t have enough confidence in the gimmick to hit the ground running with it? Who knows. What we do know is, they don’t actually do much, despite being neat little figures. If you’re into collecting things, by all means these are some of the best Nintendo-themed trinkets money can currently buy (for a generally affordable price, no less). Otherwise, if you want to see what NFC figure gaming is all about, those Disney Infinity figures are super nice, and do a lot more for you. You can also, like, find all of them in stores.

Wednesday, May 6, 2015

European Union commission says E-commerce (PC games sales) eroding national markets

The gradual erosion of national markets is one of the apparent side effects of the growth of on-line commerce and the ever increasing influence of the institutions and law of the European Union (of course with Grexit and Brexit this centralising tendency may not continue). More specifically the practice of having differential prices across Europe and the related practice of geo-blocking (preventing persons located outside a certain territory from accessing websites in another territory) is coming under increased scrutiny. This is a highly relevant issue for the games industry, given that games are increasingly being sold online and in digital form, and with a specific European Commission investigation into the pricing of games. Current developments include:
  • Reports that the European Commission is investigating on competition law grounds "the alleged geo-blocking of certain video games sold online for personal computers."
  • An investigation, again on competition law grounds, into cross border restrictions in the EU of pay TV services offered by US film studios.
  • The European Commission Digital Single Market Strategy due to be launched in May, which is likely to result in the development of digital market policy and possibly legislation.
  • A European Commission Competition 'Sector Inquiry' into the E-commerce sector which is expected to conclude mid-2016.
  • Raids on producers of electronic products and small domestic appliances, based at least in part on alleged infringements of competition law in relation to on-line sales.
It is apparent that the majority of these investigations involve competition law. There can be very significant implications for any companies that have been found to have breached competition law (including fines of up to 10% of worldwide turnover). However, the wider issues could have significant implications for the games industry as a whole- with fundamental implications for how and where games are sold, the price at which they are sold and therefore the money available to develop further games. This article considers the current position in relation to the law and thoughts on the potential outcomes of the current investigations
So what is the current position with respect to the application of competition law in this context? As you might expect a lawyer to say, the position is complicated, nuanced and ultimately should be the subject of detailed and specific advice. But looking at the position broadly the following principles are relevant.
"companies cannot make agreements with third parties that have the object or effect of restricting competition"
Firstly, of some comfort to businesses, competition law acknowledges that companies have a right to determine how they conduct their own business, who they choose to deal with and the price at which they deal. However that right is not absolute and there are exceptions. Where companies have market power over competitors and customers (referred to in competition law terms as 'dominance') they have a special responsibility not to abuse that power. This can in certain circumstances require companies to deal with third parties, or not to discriminate between customers on illegitimate or arbitrary grounds (nationality is likely to be one such illegitimate ground). Typically dominance will only arise with market shares well above 30%.
However while competition law will only rarely intervene in the conduct of a business acting on its own, companies are more restricted when it comes to the relationships with third parties. More specifically, companies cannot make agreements with third parties that have the object or effect of restricting competition. Of particular relevance to this article is the law around territorial restrictions in agreements between companies.
While it is generally deemed acceptable for a manufacturer to determine how its products are actively marketed by its distributors or resellers by allotting distributors or resellers certain exclusive territories (so called active selling), it is generally not allowed for a manufacturer to prevent those same distributors or resellers from meeting unsolicited requests for product from customers outside of their territory (so called passive selling). In this way the authorities with responsibility for enforcing competition law seek to balance the pro-competitive benefits of rationalising the distribution of products within territories, with the overarching EU goal of ensuring as far as possible that there is a single market for goods and services within the EU.
"a games manufacturer in the UK can appoint an exclusive reseller to market and sell their games in Poland and prevent them actively targeting customers the UK"
So by way of example a games manufacturer in the UK can appoint an exclusive reseller to market and sell their games in Poland and prevent them actively targeting customers the UK. However, the reseller in Poland cannot typically be prevented from fulfilling unsolicited orders from gamers in the UK.
Perhaps the best example of enforcement of this principle in practice is in relation to Nintendo. In October 2002 Nintendo and its exclusive distributors were fined a total of €167.8 million (one of the highest fines for infringement of competition law at the time) for having in place general prohibitions in its distribution agreements on the trade of its consoles and games between EU member states.
Because the European Commission sees the internet as generally a passive sales channel- any absolute requirement on third party resellers not to fulfil orders from outside the territory (which is likely to include a requirement to geo-block) is likely to fall foul of the law outlined above.
There are a number of cases other than Nintendo that follow current law and illustrate that this is more than just a theoretical risk. For example:
  • In 2007 objections were raised to Apple iTunes pricing practices which were resolved by a settlement with the European Commission based on a decision by Apple to equalise prices across the EU (or more specifically reduce the price for songs in the UK). In its statement on settlement the European Commission suggests there may not have been grounds for formal action given that this was a unilateral action by Apple "The Commission's antitrust proceedings have also clarified that it is not agreements between Apple and the major record companies which determine how the iTunes store is organised in Europe. Consequently, the Commission does not intend to take further action in this case."
  • In the much publicised Murphy case in 2012 the Court of Justice of the European Union held that the use of foreign decoder cards to broadcast FAPL football matches to a pub in the UK (thus avoiding Sky's licence fee) could not be prevented on the basis that obligations not to provide decoders outside the territory are contrary to competition law. The effect of this case was not however as dramatic as it might have been, given the finding of the courts that broadcasting to the public without an appropriate license could infringe intellectual property rights.
But competition law is not the only relevant law in this area. General provisions of the Treaty on the Functioning of the European Union (TFEU), and Directives also cover these areas - for example Article 26 TFEU states that "The Union shall adopt measures with the aim of establishing or ensuring the function of the internal market....the internal market shall comprise an area without internal frontiers from which a free movement of goods, persons, services and capital is ensured". The Services Directive of the European Parliament states that "Member States shall ensure that the recipient [of a service] is not made subject to discriminatory requirements... but without precluding the possibility of providing for differences in the conditions of access where those differences that are directly justified by objective criteria".
"There is therefore a real possibility that the outcomes of the various current enforcement actions and enquiries will be great centralization of prices for games (and other products) across the EU"
As the Nintendo, Apple and Murphy cases demonstrate there is little tolerance at a European level for cross border restrictions. That is also consistent with the broader aims as outlined in the TFEU and the Services Directive, as well as those that wish to create a single EU economy. There is also a personal element to some of these concerns - Margrethe Vestager - Commissioner for Competition EU has said "I, for one, cannot understand why I can watch my favourite Danish channels on my tablet in Copenhagen - a service I paid for - but I can't when I am in Brussels."
There is therefore a real possibility that the outcomes of the various current enforcement actions and enquiries will be great centralization of prices for games (and other products) across the EU.
However there are clearly rational objections to any single market approach. Some differences in prices across the EU may be rational however- and not simply on the basis of different costs of distribution. In a digital world the cost price differences between prices in the sale of games are not likely to be significant. But discrimination between member states can be pro-competitive in ensuring that the price of the games matches the customer's ability to pay. A single price across the EU risks that games will be set at prices that are relatively cheap in North and Western Europe, and relatively expensive in South and Eastern Europe. Ultimately this could lead to respectively oversupply and undersupply depending on the area, and may limit the profits available to reinvest in producing better games and games technology. Others complain that the move towards a single market will undermine the cultural heritage across Europe, favouring instead larger companies with pan-European operations.
It remains to be seen what the outcome of the enquiries and the investigations will be. This is however something the games industry participants should monitor carefully and feed into with their views when they have the chance. The outcome of the current inquiries could have a profound impact on the way games and other digital content is market, sold and consumed in the EU.