Internet Speed Tests

Answering a question in the SL Forum Answers, I found that I needed to update an article I wrote on testing your Internet connection to Second Life™. It is creatively titled: Troubleshoot Your #SL Connection. I posted it in October 2011.

At the time SpeedTest.net allowed one to test to a destination of their choice. SpeedTest.net has changed. Now one can only test to a nearby server from a list provided by SpeedTest.net. I suppose the idea is one only need test from your computer to your ISP. I suppose that works for most people. For our testing purpose, it is now useless.

Speed Test – Image by Lunchbox Larry @ Flickr

The article is about specifically testing your connection to an SL server. So, I need a new speed test. I started looking for a good choice to add into the article. I came across LifeWire, which has a list of speed test sites. It shows as having updated 1m/2d/2018.

Those freaking out about being splashed with freedom from the repeal of Obama’s socialist pretend Net Neutrality, you’ll need these sites to see if your connection is being throttled.

From reading the page you’ll see the main interest is in testing the speed your ISP is providing. The question they are answering is, “Are you getting what you are paying for?” So, these sites do not really answer the question we are asking, “How good is my connection to an SL server?”

There is no way to actually get these tests to download a file from an SL region server. But, you can probably find a server in the city where the SL region servers live. Pulling the geolocation of a region server from the viewer’s HELP->ABOUT… sim10446.agni.lindenlab.com (216.82.51.152:13021) I get San Jose, California.

A site named BandwidthPlace.com will let me select a server to test to. That is as good a bandwidth test to SL as we are going to get. You’ll have to check to see which servers your viewer is using, geolocate them, and then find a testing service with a server in that city.

You may have to work through the LifeWire list to find a service that will let you test to the city you want.

Hardware: Picking a New CPU for Second Life

The skeleton and brain of a computer is the motherboard and CPU.  A CPU is the Central Processing Unit. This is most like our brain.

The mother or main board is like out nervous system or skeleton. It is the large circuit board that every part of the computer plugs into, including the CPU. It holds everything together.

For detailed information on CPU’s see the video: (jump to TM 6:07 to skip the explanations and see the benchmarks.

There is a lot of jargon in the video, I’ll get the important stuff explained.

First the TL:DR, the video shows there is very little difference between the various CPU’s. When it comes to frames per second. The three tested CPU’s are about equal. However, if you are going to use Intel’s HD Graphics without a dedicated graphics card, he shows there is a difference and the 6th generation CPU wins. So, laptop people… go for 6th gen CPU or a dedicated video chip. Continue reading

Hardware: Data Storage – What is the Best Buy?

As I built my new computer and researched hardware I realized I was falling behind in my technical knowledge. I had to sort out a number of things to be able to intelligently decide which was best technically and for the money. This article is on the aspects of storage devices and proves a lot of information on SSD’s.

SSD M.2 Connectors

SSD M.2 Connectors

I wrote Hardware: Disk Performance Compared to provide just performance information. I was concerned with performance of Second Life™ an OpenGL based game/virtual world. Here I’ll get into the complications behind getting good performance, cover how to make comparisons for the best buy, and provide information on how it affects Second Life performance.

Numbers

To get the most for your money you are going to have to be able to see through the advertising hype. So, as you read specs and try to make comparisons you’ll run into a variety of designators for data transfer rates and storage volume. They are all worded to make a product sound good.

Let’s start with the upper case ‘B’ and the lower case ‘b’, which have specific meanings. In the simplest use 1B = 8b, or 1 byte equals 8 bits. A bit is a 0 or 1. A byte is a total of eight 0’s and 1’s. (For more on this see Online Conversion, a tutorial.)

In some places we see 6GBps used to mean 600Mbps. This can look like B = 10b. Here because they are talking about data moving from storage devices through the transfer path they add in the data overhead of 2b (bits) for error correction and 1B does equal 10b in THEIR math. They are taking some liberties in their ad copy. One often has to CAREFULLY look at the numbers and determine the context intended to understand the size of a ‘byte’ being discussed.

In some ways the B’s, G’s, and M’s are misleading in how they are used. It is easy to make a case for that being deliberate. The actual data being moved is 8b/10b = 0.8 * 6000Mb raw data = 4,800Mbps actual usable by you data. But, manufactures like the bigger number and they are literally moving 6Gbps. It just isn’t all your data that is getting reported. They assume the knowledgeable user will infer their true meaning. I suspect they know a whole bunch of people won’t.

Unless you are comparing devices and protocols on a way technical level thinking in terms of data rates as 6GBps verses 4.8GBps doesn’t matter much. Consider. Is 6GBps SATA III faster than 3GBps SATA II? Yes. Is 4.8GBps faster than 2.4GBps? Yes. The difference wouldn’t change the take away that SATA III is faster than SATA II regardless of which numbers you use. Nor when comparing prices does it really change what we decide.

The take away here is to be sure you use the same measures for your comparisons. Also understand these are the THEORETICAL values of maximum performance which is seldom achieved.

More pages, links below…

HTTP Concerns

My readers know the HTTP aspects of Second Life™ are being upgraded to the latest and greatest tech available for HTTP. For most of us this is a non-issue we hope will give us better overall performance; faster texture download, more stable connections, and world that renders quickly.

ADITI Regions

ADITI Regions

For others this is a matter of their in-world systems, think vending machines and subscription machines, working or not. If you use vending machines or other tools that depend on a third party server, you should be checking with the creator to see if they are testing their products with the new HTTP protocol services.  Continue reading

Second Life Inventory Beta Tests

Yesterday in the Server User Group Jenn Linden attended to request help with testing coming inventory changes. These tests are in progress now on the ADITI grid. This set of changes is about speed and reliability. This set of changes is not part of the Direct Delivery (DD)feature. It is expected to improve DD’s performance too, but the change is for improved speed and reliability.

Want to Help and Test?

The code for the change is deployed on the ADITI grid. Instructions for the test can be found in the SL Wiki. See Inventory Beta Tests. The page includes the regions names of the areas using the new inventory code.

It helps if you take someone with you to test or make a friend while there. Part of the test is about giving inventory items to others and receiving items from others.

There is also an odd step where you are asked to: Remove an attachment such that it tries but cannot rez and must be returned. That could have been worded better. It sounds a bit over educated. Basically, you need to DROP and attachment in a no-rez area. Oskar Linden set up some no-rez areas in the test regions for that part of the test. Pay attention to parcel names within the regions.
Continue reading