

Metacritic: User Reviews Are Starting!
#741
Posted 19 September 2013 - 12:26 PM
#742
Posted 19 September 2013 - 12:29 PM
Nekki Basara, on 19 September 2013 - 12:26 PM, said:
Since launch. The weeks and months prior to that you'd see anywhere from 80-200 people watching the stream. Now you see 300-600 (saw 681 at one point) watching over the past few days.
#743
Posted 19 September 2013 - 12:31 PM
#744
Posted 19 September 2013 - 12:49 PM
Heffay, on 19 September 2013 - 12:29 PM, said:
Since launch. The weeks and months prior to that you'd see anywhere from 80-200 people watching the stream. Now you see 300-600 (saw 681 at one point) watching over the past few days.
Nekki Basara, on 19 September 2013 - 12:31 PM, said:
Y'all are restoring my faith in humanity asking for concrete information, but then I was really blown away when he came back and provided it.
#745
Posted 19 September 2013 - 12:54 PM
#746
Posted 19 September 2013 - 12:57 PM
#747
Posted 19 September 2013 - 01:00 PM
Kurshuk, on 19 September 2013 - 12:49 PM, said:
Y'all are restoring my faith in humanity asking for concrete information, but then I was really blown away when he came back and provided it.
Nekki Basara, on 19 September 2013 - 12:54 PM, said:
Have you guys ever heard of ghost machines?
I hate to break it to you, but one decent gaming rig can ghost 60 stream hits at a time with little to no effort.
But hey, paid shills would never stoop so low as something like that, they did not after all call for positive vote bombing on metacritic, right?
Edited by Franchi, 19 September 2013 - 01:01 PM.
#748
Posted 19 September 2013 - 01:00 PM
Heffay, on 19 September 2013 - 12:15 PM, said:
That's great. They must've seen someone grouped on NGNG get themselves annihilated while there was this lone hunchy (me) that was killed last because it was instantly 12v8. Thanks guys for communicating.
#749
Posted 19 September 2013 - 01:03 PM
Nekki Basara, on 19 September 2013 - 12:54 PM, said:
The "people watching the stream" value provided by Twitch.tv?
Or are you implying that I'm making it up?
Franchi, on 19 September 2013 - 01:00 PM, said:
I hate to break it to you, but one decent gaming rig can ghost 60 stream hits at a time with little to no effort.
But hey, paid shills would never stoop so low as something like that, they did not after all call for positive vote bombing on metacritic, right?
Right. They set up an ESX server farm during launch to stream multiple connections and made it seem like there are a lot more people watching.

#750
Posted 19 September 2013 - 01:08 PM
Heffay, on 19 September 2013 - 01:03 PM, said:
The "people watching the stream" value provided by Twitch.tv?
Or are you implying that I'm making it up?
#751
Posted 19 September 2013 - 01:10 PM
Nekki Basara, on 19 September 2013 - 01:08 PM, said:
Ok. Suit yourself. I know what I saw. You can watch the streams and verify the current numbers yourself.
#752
Posted 19 September 2013 - 01:17 PM
#753
Posted 19 September 2013 - 01:18 PM
RG Notch, on 19 September 2013 - 07:05 AM, said:
This is exactly what I would expect from a launch that was nothing more than a bugfix and the removal of the "beta" designator.
Doesn't keep me from enjoying MWO. It's fun. If it isn't for you - boo hoo.
#754
Posted 19 September 2013 - 01:27 PM
Morbid Jester, on 19 September 2013 - 06:53 AM, said:
so they are, so I did. And with the current score sitting around 5.3 or something that means - fanboi and hater's scores included - MWO is still slightly better than average....
I have no clue what world you live in but 5.3 is not average it's actually really low
the games current score would actually put it on the sixth page of worst scores ever with the current 5.3 score being translated into 53%
http://www.gameranki...l?page=5&sort=1
#755
Posted 19 September 2013 - 01:43 PM
Toydolls, on 19 September 2013 - 01:27 PM, said:
the games current score would actually put it on the sixth page of worst scores ever with the current 5.3 score being translated into 53%
http://www.gameranki...l?page=5&sort=1
You're comparing MWO's user reviews with professional reviews on two different websites. The average User reviews are almost always lower than the average profesional reviews. I'm not even sure that those two websites use the same metrics for averaging critic reviews.
#756
Posted 19 September 2013 - 01:44 PM
Jman5, on 19 September 2013 - 01:43 PM, said:
indeed but the real reviews won't be far off you can count on that
#757
Posted 19 September 2013 - 01:49 PM
Heffay, on 19 September 2013 - 01:10 PM, said:
Ok. Suit yourself. I know what I saw. You can watch the streams and verify the current numbers yourself.
When NGNG's team speak goes from 120 to its cast at 80 it's hilarious.
Hell KONG fills their teamspeak on a whim, a whole 10-15 players outnumber its viewers.
#758
Posted 19 September 2013 - 02:14 PM
Kyynele, on 19 September 2013 - 09:51 AM, said:
Your post was mostly good, but fixing something wrong with doing more wrong doesn't lead to good. 3rd person view should've been left out, but splitting the queues would be much worse for the game than adding 3PV was. More queues leads to less players in queues, means either longer waiting times, worse matchups or some queues empty.
PGI knows this, that's why it's not in the game. Not because it'd be lots of work.
Don't you find it odd though that splitting the queues is a big issue?
Looking at that closely, what does that mean?
Is the community that small that splitting the queues is that significant an impact on match-making? If so, after the tepid launch, current "meh" reviews, and forum hostility that makes me wonder if the game is already too small, number-wise, to succeed.
If it's not the numbers, wouldn't a split have provided some valuable information/insight for the Developers?
What if statistically 3PV ran rampant over 1st PV players?
What if the opposit happened and 1PV dominated?
What if the numbers varied but were always somewhat close?
Any one of those situations would provide data, "metrics" if you will, that would supply valuable information to help in the decision making process and I shudder to think that they didn't just do it out of fear of not having enough players to support the queues.
It was Beta, and that's the time they should be jinking around with the game, collecting info on what works, and what the community likes. If it didn't work they could quickly roll back to a previous patch, but they didn't even try it.
The whole thing is "odd".
Edited by TLBFestus, 19 September 2013 - 02:15 PM.
#759
Posted 19 September 2013 - 02:30 PM
TLBFestus, on 19 September 2013 - 02:14 PM, said:
Don't you find it odd though that splitting the queues is a big issue?
Looking at that closely, what does that mean?
Is the community that small that splitting the queues is that significant an impact on match-making? If so, after the tepid launch, current "meh" reviews, and forum hostility that makes me wonder if the game is already too small, number-wise, to succeed.
If it's not the numbers, wouldn't a split have provided some valuable information/insight for the Developers?
What if statistically 3PV ran rampant over 1st PV players?
What if the opposit happened and 1PV dominated?
What if the numbers varied but were always somewhat close?
Any one of those situations would provide data, "metrics" if you will, that would supply valuable information to help in the decision making process and I shudder to think that they didn't just do it out of fear of not having enough players to support the queues.
It was Beta, and that's the time they should be jinking around with the game, collecting info on what works, and what the community likes. If it didn't work they could quickly roll back to a previous patch, but they didn't even try it.
The whole thing is "odd".
There's nothing odd about it. They don't have the playerbase to split the queues. Notice they took away the player counter. Notice they snuck the forced 3PV on us without any warning. Notice how they always mention registered people not concurrent players. Notice they talk about player retention not growth. Notice they are proud 50% of Founders log in once a month.
Put that all together it's easy to see they have an issue with player numbers.
#760
Posted 19 September 2013 - 02:54 PM
TLBFestus, on 19 September 2013 - 02:14 PM, said:
What if statistically 3PV ran rampant over 1st PV players?
What if the opposit happened and 1PV dominated?
What if the numbers varied but were always somewhat close?
What if 3PV dominated because the data was collected after "launch" when the queues were full of newbies who didn't know you could switch from the default 3PV to 1PV, and PGI declared this proved that they were right all along and that everyone wants 3PV?
...but of course the guys who declared we the vocal minority are on an island would never do something like that, would they?

8 user(s) are reading this topic
0 members, 8 guests, 0 anonymous users