New Information on Alpha Testers

[hrdndv]

The following is from the Japanese forums. I have tried to clean up the translation but it is far from perfect. SE said an English version will be posted soon.


Naoki Yoshida said today (11/27/12) at 1:08 AM:
"Since we are developing a technique that divides the ongoing development, called the branch,
skinny β test work is already progressing on a branch separate from the α. The α test will be extended by a period of time, though not as the only test, before the world α. (Closure of the hardware test does not interfere with the construction of the world as it is also a β test). Alpha testing is steady Therefore, whether the β test can say ahead of schedule, is a different matter."

Future plans for α test
I am Yoshida producer / director. Thanks for waiting. We will report plans for the future of newborn FFXIVα test, . However, isince there is no description of the contents of the α test in the first place, and because I think the intent is not transmitted correctly, please read the following (This is long sorry). As has been reported in a previous letter from the producer, the α test is only a side management / development "stress test" of the server and can not be used ​​for the purpose of a MMORPG. [ie it is not a game] However, I think the load test itself has a sort of image that is not clear, let me explain briefly in the next section

Overview and load test server configuration nascent FFXIV per 1 World, FFXIV newborn has been designed as an average peak number of concurrent connections at 5000. This number is the number of simultaneous connections to the world that we can create easily. A density of 5,000 simultaneous accounts at the peak time of the play is the maximum number of people to be accommodated in one world at the same time. As I recall, because it is "Capacity simultaneous whole" per world, the one world of FFXIV newborn consists of around 40 zones, and it is assumed that our players must be logged-distributed to each zone. Since the server to accommodate 5,000 people in 40 zones, ie a capacity of one zone server, the total will be much lower than 5000. As character level rises, they are distributed to the content, from the width of playing style. This means that all players tend to gather in one zone, which is virtually impossible. (... But it is is possible if I create the situation deliberately) . FFXIV newborn, has been designed to accommodate possibly 1,000 people in one zone, still, "Simply put 1,000 people in one zone In order to test! " is okay. So you can perform α tests, and you can change both.. (Has been confirmed in the content and the conductor, with more than 1000 characters, whether alone or a collection. Additionally, [this holds when players were] instantiated for each Zone). We validated this load for battles, several at the same time, and this has been done with some gathering or at the same time. Whether monitoring of running some quests, the only situation in which there are a lot of people in the zone server we simply have not verified yet. ...... Whether or not there is a problem with transfer zones has also been studied at the same time with 300 to 400 people. Neither cause any problems during intensive access to the patch server and does not produce a client crash on specific hardware, or control POP / DEPOP monster when it is working correctly in the event of high load , etc. We are a large-scale validation will help us to everyone on the tester.

Why a lot of people can not participate in the α test? ■ It is assumed that the test is fo the purpose of testing the server load. When viewed from our players standpoint, α test's playable range is extremely small compared with the retail version. Graphics are normally rough. There are many temporary implementations of the UI, in many cases not even involving a quest scenario. So we only perform a large number of simultaneous battles. Also, play range is limited as follows in α test of FFXIV newborn: · Races that can be used include only woman Mikotte / Sunseeker man and Hyuran / Mid Lander only supports creating new characters for everyone, and cap at the class level each 20-. Only Guridania can be used as the starting country. Professional Sojutsu, Battle Classes Archer Only certified genjutsu -Crafter / Gyazara is Carpenter, leatherworker, only master gardening Opening is tentative only α test unimplemented cut scenes (will be displayed black mask) only mausoleum Tamutara · Dungeon (Lv15 -Lv20) for ) those of the middle-conductor Quest implementation exists beyond (Lv10 to Lv1 ~ Lv10 -Girudorivu the production Reeve (is) to Lv20 to Lv15
I think I can play very well in a percentage of the α test, But to play zone, mainly focused on about 4 zone (open range 8 zones). Because it was created though quite conductors initially, FFXIV nascent login by thethousands simultaneously and can only start in one city, when you start to play, along with the level up, so that the high load is always applied to the same zone. Further, since there is no content in other areas, even reaching the Lv20 cap, this means they tend to accumulate in the initial zone. Because the original purpose is server load test and the purpose is to get the load in the targeted zone, so as α test, it is correct to do this. However, because it is calculated at 1000, the maximum load per zone is as I mentioned earlier, and by simply increasing the number of players, and extending those same increases in addition to the players remaining in the zone, results in extreme loads that I can not happen under normal playing conditions, I would not be able to play decently at these loads. Even if we elected to test long-awaited α [test 4], I still can not play satisfactorily. So the only solution is to not only increase the number of α test worlds, but in addition, restrict α-play test range and conventional character data as well. Because at the end of the wipe test α (erased), the rate tends to show decreased logins. So, in order to distribute accounts, we will log in people who are new testers, to continue the validation as is customary. We expect it to be at the steering nascent, with respect to the number of distribution accounts, the rate of login and will be approximately 20%, which I had predicted, the rate of log remained at less than -50% 40% still at present Because you are, it is a situation that while calculating the load, in order to increase the test account

The progress of the test α enters the third of the four phases of all phases from newborn FFXIV α test tomorrow. The first phase of operation one world foundation load test. Load test proceeds very smoothly thanks to all testers, 850 simultaneous connections, at about 270-280 per concurrent battle zone, at this point it is possible to find the critical load or bugs and I could fix them immediately. Further, a process prone to high-load test was performed followed by simultaneous tests, I have reduced processing speed in detail. In the test phase α 2 we have an increase in testers opening the second world. I tested simultaneous operations in multiple worlds and found there are almost no noticeable problems.
Fortunately, support and faster processing, during which time I have a final check on the parallel start of the global test in Britain, Germany and France. Endurance test is 3-phase, long-term management of the global operational test + I starts tomorrow. In accordance with the Phase 3, 11/26, are distributed to all regions also add a test accounts. Subsequent Phase 4, slated to be carried out, such as integrated test world, while looking at the state of Phase 3, the schedule here is we're going to decide. However, since the strong request of you, until near the end of the year, as much as possible (we are not what you would call a α test anymore ...) that you are considering to release the α test world.


the progress of the test α enters the third of the four phases of all phases from newborn FFXIV α test tomorrow. The first phase of operation one world foundation load test. Load test proceeds very smoothly thanks to all testers, 850 simultaneous connections, at about 270-280 per concurrent battle zone, at this point it is possible to find the critical load or bugs and I could fix them immediately. Further, a process prone to high-load test was performed followed by simultaneous tests, I have reduced processing speed in detail. In the test phase α 2 we have an increase in testers opening the second world. I tested simultaneous operations in multiple worlds and found there are almost no noticeable problems.
Fortunately, support and faster processing, during which time I have a final check on the parallel start of the global test in Britain, Germany and France. Endurance test is 3-phase, long-term management of the global operational test + I starts tomorrow. In accordance with the Phase 3, 11/26, are distributed to all regions also add a test accounts. Subsequent Phase 4, slated to be carried out, such as integrated test world, while looking at the state of Phase 3, the schedule here is we're going to decide. However, since the strong request of you, until near the end of the year, as much as possible (we are not what you would call a α test anymore ...) that you are considering to release the α test world

Yoshi explained the selection process in detail in Japanese. Basically, the current alpha test program is very small so that everyone who is invited tends to pile up in the starter area with no place to go. This puts strains on the program so that they have a limit of 1000 players in a server. Each time you sign out your character data is lost so you have to create a character every time you sign in. This puts a limit on the number of people that can sign in at one time. They refer to this as "sequentially distribute account. ... To continue the validation is customary."
"Phase 3 tomorrow (11 / 26), operational endurance test starts global operational test + for a long duration test, according to phase 3 all region test account."
"After phase 4 begins, world integration testing will begin."

■ number of Alpha test account distribution and future distribution plans ■

"Please note that we cannot disclose an exact figure of the current alpha test email deliveries.
Japan, in 11 / 27, was 1/4 of the total number of the applications recieved emails.
For North America's global test, 1/6 of the total applications recieved emails.
For EU, approximately 1/3 of the total applications recieved emails.

There are additional plans for future stability testing of phase 3 until December 2, 2012. We will add more Tester on December 3, 2012.
Also, initially there is no planned tester load. However, [these considerations] will allow me to open the third world for alpha testing after December 3, 2012.

We will add to the Testers from time to time while watching login rates begining December 2-3, 2012 or later,.
Concerning NS ... we will add thousands each week, but will be watching how busy the servers are to decide. Please note that."

Ask about the rate of increase in α test accounts
As well as FFXIV new team, the fact that steering the end of the current period needs everyone, I strongly believe that people who have submitted applications, and I want to pass these accounts as many people as possible, I hope that those who elected to test it, as much as possible, will join us to play the second world, The third in the world for the α test I think I can ask. For the convenience of the open range α test, if the players concentrate on the World as described above, then this results in a load can not occur in operation. Usually calculated increases in test accounts, suggest you are no longer forced to be cautious. "Although we are currently operating in the first two worlds called," Chocobo " and " Moogle". I think as far as possible, "If it is possible to play in creating a character in "Moogle". It would be very helpful if whoever joins the α test after 12/3 would choose the third world that was added. I will be over! (As soon as it is possible to translate this letter it will be. I wrote this letter in a hurry!)


A further note states that they intended to add more alpha testers but the alpha version is so good that noone wants to leave!