1911 replies to this topic

[Alaskan Nobody]

Cimarb, on 11 June 2014 - 10:29 AM, said:

Why can't everyone wait until we see it in action before asking for it to be turned off?....

Because then what would they whine about?

Having said that - there probably will be a toggle....eventually

[CeeKay Boques]

Cimarb, on 11 June 2014 - 10:29 AM, said:

Why can't everyone wait until we see it in action before asking for it to be turned off?....

Because the sky is falling, duh... you wanna get crushed under its weight?

[Mawai]

Cimarb, on 11 June 2014 - 10:29 AM, said:

Why can't everyone wait until we see it in action before asking for it to be turned off?....

LOL ... well lets see ... I already know what the HM sound is like ... it is annoying ... at least to me ... I also tend to play with the sound turned down due to other people in the house ... having a large number of mechs on the field equipped with a variety of sounds like "Sonic Boom" and others ... can't be any better than a couple of lances of Heavy Metal ... and it is pretty easy to judge based on experience that I would appreciate the ability to toggle this particular sound effect on and off.

The game already has sliders for effect volume, dialog volume, music volume and a checkbox for Frontend Sound in the mechlab ... is it that much of a stretch to request a checkbox to turn on/off warhorns?

I have nothing against the idea or how they choose to implement it but I really don't want to bombarded by tunes and noises when a mech scores a kill.

Gizmoh, on 11 June 2014 - 09:49 AM, said:

I'd also would like an option to disable the in-cockpit visual of the warhorns, some of them are quite jarring.

Well ... I think if you choose to equip one you should be stuck with the look ... but I don't think everyone else in the game should have to listen to it

Edited by Mawai, 11 June 2014 - 11:04 AM.

[Cimarb]

I really don't think it will be as bad as some people think, especially after a couple weeks. Much like a new mech, everyone will be using one for the first week or two, but after the novelty wears off, most people will get tired of them and unequip them. By the time a toggle is in place, I don't think it will be an issue.

[Imperius]

Karl Berg, on 09 June 2014 - 11:10 AM, said:

Jin was definitely working on this very recently in fact. Just talked to him, and it seems DX11 is now nicely fixed up; but he's waiting on some responses from nVidia, and potentially some driver fixes for DX9 before it's releasable.

The new 337.5 nVidia drivers apparently have some excellent optimizations for MWO's DX11 path. Those drivers are still in beta, but apparently can have a very large impact on D3D CPU costs and game performance.



Interesting. I'll escalate this as well. It sounds like it's just slipped through the cracks.

edit: Indeed an oversight. QA is reopening and reassigning the bug now.

Karl Berg I have a very important question I would like to ask.

What would be the best PC to build to play MWO at optimal FPS at Very High Settings? I am looking to build a game rig on the cheap and I was wondering what you guys/girls use to play and test the game with. I'm also looking to play it at a resolution no bigger than 1080p. Also I would like it to be able to run those 2-4k texture files you may release in the future.

i7 or i5
RAM Speed (Does it matter to MWO)
SSD or HDD

I have an Nvidia GTX 680 but I think my dual 2.8 Xeons(quad-core) are bottlenecking it's potential. It's a 2008 pc so its time to update I guess.

[CompproB237]

Imperius, on 11 June 2014 - 09:20 PM, said:

What would be the best PC to build to play MWO at optimal FPS at Very High Settings?

Not much you can do. The client isn't very well optimized.

[Mawai]

Imperius, on 11 June 2014 - 09:20 PM, said:

Karl Berg I have a very important question I would like to ask.

What would be the best PC to build to play MWO at optimal FPS at Very High Settings? I am looking to build a game rig on the cheap and I was wondering what you guys/girls use to play and test the game with. I'm also looking to play it at a resolution no bigger than 1080p. Also I would like it to be able to run those 2-4k texture files you may release in the future.

i7 or i5
RAM Speed (Does it matter to MWO)
SSD or HDD

I have an Nvidia GTX 680 but I think my dual 2.8 Xeons(quad-core) are bottlenecking it's potential. It's a 2008 pc so its time to update I guess.

Although I know you would like Karl's input ... this is actually a question that is often asked and answered by the community.

There is even a forum for this specific topic and I have seen similar questions asked and answered there so it would be a good place to look first.

http://mwomercs.com/...re-accessories/

On your actual question ... I use an older AMD Phenom II x4 965 with an AMD Radeon 6970 and get 30 to 40 fps on very high most of the time. It isn't perfect but it is very playable.

The best performance you will get would be with the fastest hardware ... i7-4770k+ AMD 290X or the equivalent Nvidia card + 8 to 16Gb of RAM ... SSD may help load times but it should not affect game play ... MWO doesn't yet have SLI or Crossfire support (though I am in the camp that you are generally better off with a faster single card that two in SLI/Crossfire ... though it may be a viable upgrade path in a couple of years).

However, the difficult question to answer is "What is acceptable performace and how much do I want to pay for it?" This is where the i5 vs i7 question comes in ... is an i5 any slower? If it is 2% slower is that worth the cost increment to an i7? 5%, 10%? In addition, the other hardware, drivers, other software installed and running ... can have a significant impact on performance ... so even systems with identical hardware may have somewhat different performance. So, although the question you ask seems really simple ... it has more aspects to it than it would first appear. So your best bet is probably to look at the stated game hardware requirements and look at the information in the forum I linked above.

(Memory speed is a more significant issue for SOC solutions with integrated graphics that use the system memory for the video card. In this case the memory speed can have a significant impact on gaming performance. On the other hand, for typical applications with a discrete graphics card with its own fast memory then the speed of the system RAM should be less of an issue.)

Edited by Mawai, 12 June 2014 - 06:00 AM.

[Mawai]

Cimarb, on 11 June 2014 - 08:23 AM, said:

You are correct. If you continue to beat your dog even after it stops peeing in the house, because you are holding a grudge for the pee issue, then you have now reached the point I am talking about! In this case, the "dog" has been beaten so many times, it would rather just stay outside than come in near you/us.

Interesting analogy ... but people are not dogs ... we are supposed to be adults and communicate. Karl does an excellent job ... he acknowledges feedback, both positive and negative, and is up front about stating what he can say and what he can't.

The only reason that PGI has a negative reputation is because of their failure to communicate.

Let's look at some examples of design decisions with poor communication:

1) double vs. 1.4 heat sinks - PGI justification for this decision - "A jenner could core an Atlas from behind in 4 seconds if we used double heat sinks" - this statement was easily shown to be incorrect both mathematically and in practice (especially since engine heat sinks are true doubles already ... most Jenners fit a 300XL ... and already have 10 inherent true double heat sinks ... and don't have the tonnage to fit many more) ... the change in cooling for a Jenner is inconsequential.

2) Heat scale design: in previous incarnations the heat sinks increase cooling not capacity while in MWO they do both. Add a heat sink and you increase the maximum as well as cooling rate. I've never seen a decent explanation for this one.

3) ECM: ECM was a major bone of contention in the community. It was and is the most powerful 2 slot 1.5 tons you can use in the game. When introduced it had no counters, totally changed the balance of the game, and generally shifted the entire game in a new direction. It worked as a perfect hard counter to streak cats ... (though the issue there was bugs with the streaks hitting the CT almost exclusively). I have never seen a post by PGI explaining the reasoning behind the ECM design decisions. PGI simply ignored the feedback in the forums, many many many reasonable comments and suggestions were made long before the discussions became vitriolic. The only reason the discussion became nasty was due to NO communication or indications that PGI might actually be listening to the feedback. (Personally, I still think ECM is not well designed but the odds of it being changed at this point are really small).

4) LRMageddon. PGI changed the arc tracking of LRMs so that the last targeting changes were closer to the end of the arc. This allowed missiles to virtually plummet onto mechs ... there was no terrain significant enough to hide behind and LRMs were also preferentially hitting CT. I can't recall how many days it took for PGI to hotfix it ... but this was one of many changes that should never have hit the live servers.

There are so many decisions made by PGI that have ignored comments and player feedback that it is hard to count. PGI has stated that the forum community is not representative of the player base, they have made comments that make it sound like they know better and that they have no responsibility or requirement to respond to folks on the forums. Honestly, if you want to know who the dog is in this scenario ... I would have to suggest it is the community on the forums ... given the apparently paternalistic attitude and "we know best" approach taken by PGI and the total lack of respect they seem to have offered to their most ardent fans from the very beginning ... is it surprising that they are not greeted with open arms?

Do you recall Russ' justification for ending closed beta and going to open beta? "Out community is tired of testing, they are burning out, they have nothing further to offer in the context of a closed beta ..." ... that has to be among the most self-serving tripe I have ever heard ... they received the feedback from the closed beta community and did not like it ... they had plans to go to open beta and start the revenue stream going ... they could have stated that up front but instead decided to justify their decisions as being in the interests of the the community despite comments from the community that significant game functionality was missing or subpar at the time the game went to open beta. At least they delayed it to October to try to fix some of the biggest issues.

Sigh ... sorry for the rant ... the point is that PGI has been bad at communicating since closed beta. The community does not respond negatively because PGI communicates ... they respond negatively because they don't and have an on-going and continuing history of not responding. Just look at the clans - they are supposed to release in 5 days - there are no numbers on any of the clan weapons ... just some concept write ups in command chair posts. Damage/heat/ROF, tonnage, crit slots, all of this goes live in 5 days but none of the information is available. Some of it will be available if you participate in todays test but honestly ... tentative numbers should have been released at least a month ago so that folks have an idea of what is coming and an opportunity to offer feedback.

[Draconis Nagasawa]

Mawai, on 12 June 2014 - 06:33 AM, said:

Sigh ... sorry for the rant ... the point is that PGI has been bad at communicating since closed beta. The community does not respond negatively because PGI communicates ... they respond negatively because they don't and have an on-going and continuing history of not responding. Just look at the clans - they are supposed to release in 5 days - there are no numbers on any of the clan weapons ... just some concept write ups in command chair posts. Damage/heat/ROF, tonnage, crit slots, all of this goes live in 5 days but none of the information is available. Some of it will be available if you participate in todays test but honestly ... tentative numbers should have been released at least a month ago so that folks have an idea of what is coming and an opportunity to offer feedback.

Non disclosure agreements and other legal jargin that people tend to forget. While PGI is the train, it is IGP that drives it. There are plenty of things that PGI can not do (this has been explained previously).

You also say that there is no communication (which is subjective to an individual), yet there is 7 posts by PGI within the week about future content.

Reminding everyone again, that PGI is small when compared to Blizzard or SOE. Do you really want them to respond to every needy person to full fill every single question? Or do you want them to finish the content?
There are not enough people for that, let the developers do their jobs, poor PGI is always under the microscrope which is not neccesary.

Edited by Crusadiar, 12 June 2014 - 08:37 AM.

[Imperius]

Mawai, on 12 June 2014 - 05:55 AM, said:

Although I know you would like Karl's input ... this is actually a question that is often asked and answered by the community.

There is even a forum for this specific topic and I have seen similar questions asked and answered there so it would be a good place to look first.

http://mwomercs.com/...re-accessories/

On your actual question ... I use an older AMD Phenom II x4 965 with an AMD Radeon 6970 and get 30 to 40 fps on very high most of the time. It isn't perfect but it is very playable.

The best performance you will get would be with the fastest hardware ... i7-4770k+ AMD 290X or the equivalent Nvidia card + 8 to 16Gb of RAM ... SSD may help load times but it should not affect game play ... MWO doesn't yet have SLI or Crossfire support (though I am in the camp that you are generally better off with a faster single card that two in SLI/Crossfire ... though it may be a viable upgrade path in a couple of years).

However, the difficult question to answer is "What is acceptable performace and how much do I want to pay for it?" This is where the i5 vs i7 question comes in ... is an i5 any slower? If it is 2% slower is that worth the cost increment to an i7? 5%, 10%? In addition, the other hardware, drivers, other software installed and running ... can have a significant impact on performance ... so even systems with identical hardware may have somewhat different performance. So, although the question you ask seems really simple ... it has more aspects to it than it would first appear. So your best bet is probably to look at the stated game hardware requirements and look at the information in the forum I linked above.

(Memory speed is a more significant issue for SOC solutions with integrated graphics that use the system memory for the video card. In this case the memory speed can have a significant impact on gaming performance. On the other hand, for typical applications with a discrete graphics card with its own fast memory then the speed of the system RAM should be less of an issue.)

Thanks for the reply. I will look though that thread but I still hope to get an answer from a dev about what system the really build for. Sure ATI works but the advertise and are "powered" by Nvidia, at least CryEngine is anyway. Second yes I'd rather get an i5 as more people have gotten better game performance out of them than i7s due to developers tweaking games toward them because they are the "most popular" CPU for most peoples budget builds. MWO is pretty much the only PC game I play, I really don't plan on upgrading the GPU a GTX 680 with 4gb of GDDR5 should run the game fine, but I think the issue is bottlenecks due to the old MB and Xeons. They are great for rendering in graphic programs like Adobe. It comes to games they just aren't optimized for them cause very few "gamers" use those CPU's. I don't know a thing about CryEngine or what Hardware makes a difference in MWO.

My current rig doesn't do too bad but I have friends that get "better" performance than me with older cards. It doesn't seem to matter what my graphic settings are so it must be a bottleneck with the MB or the CPU... It lasted this long and will keep chugging on but it's time to build a budget PC that can run this game in it's prettiest form on 1080p. I don't plan on making the 4k jump for at least 3-5 yrs.

I'm not asking specifically to build a pc for me just would like to know what they plan to "optimize" for hardware wise.

I could go blow 3k on an Nvidia Titan Z and watch it do nothing... I know better than that. I would have bought a 5-6k rig this year it decided to get a 2014 Impala LTZ instead , better purchase IMO.

[p4r4g0n]

Mawai, on 12 June 2014 - 06:33 AM, said:

-snip-
There are so many decisions made by PGI that have ignored comments and player feedback that it is hard to count.
-snip-

Main reason I mostly only provide feedback on critical issues and never bothering with investing too much time with making suggestions.

Crusadiar, on 12 June 2014 - 08:14 AM, said:

-snip-
You also say that there is no communication (which is subjective to an individual), yet there is 7 posts by PGI within the week about future content.

I believe Mawai was speaking of the past. However, if I interpreted his post wrong, then I have to concur that it is much better now.

Having said that, while communications may be better now, I'm on the fence on any announcements about forthcoming features / mechs / etc till I actually see them released. Past experience with changes made between initial announcements and subsequent release & re-balancing make me extremely cautious.

[Mawai]

Crusadiar, on 12 June 2014 - 08:14 AM, said:

Non disclosure agreements and other legal jargin that people tend to forget. While PGI is the train, it is IGP that drives it. There are plenty of things that PGI can not do (this has been explained previously).

You also say that there is no communication (which is subjective to an individual), yet there is 7 posts by PGI within the week about future content.

Reminding everyone again, that PGI is small when compared to Blizzard or SOE. Do you really want them to respond to every needy person to full fill every single question? Or do you want them to finish the content?
There are not enough people for that, get over your selves.

Sigh. Sure there are lots of things that PGI can and can't do. However, in closed beta, everyone was under an NDA and PGI still didn't do a great job of communicating. In addition, information on design choices is something they could share ...certainly after it has gone live.

PGI is a small shop. ~50 people ... 8 engineers ... 3 game, 3 system and 2 UI. Number of game designers and artists .. lots more than engineers. Time need to write a brief description to satisfy hundreds? An hour a week?

Finally, whoever said anything about "to respond to every needy person to full fill (sic) every single question" ... that is both exaggeration and fabrication ... You will be hard pressed to find even one reasonable reply from PGI to many of the concerns or questions raised by the community.

One last comment, yes they released several clan related command chair posts less than a week before all these systems go live. These are somewhat short on details but any information will do at this point. However, perhaps it demonstrates a desire to be more communicative going forward ... in which case I applaud their efforts and encourage them to continue. However, lets look at the history ...

June 2014 - 8 - not surprising given the clan release
May 2014 - 4
April 2014 - 4
Mar 2014 - 1
Feb 2014 - 6
Jan 2014 - 2
Dec 2013 - 9 - clan/CW/Launch module initial updates
Nov 2013 - 3
Oct 2013 - 1
Sep 2013 - 3
Aug 2013 - 4
Jul 2013 - 4
Jun 2013 - 3
May 2013 - 2
April 2013 - 5
Mar 2013 - 7
Feb 2013 - 4
Jan 2013 - 4
Dec 2012 - 4
Nov 2012 - 6
Oct 2012 - 3

Except for the month with the initial announcements of clans/CW and launch module (which they still have not been able to deliver on - though I know Karl is working hard on it ) ... this month is the most command chair posts ever released. So, I don't think you can hold it up as an example of PGI's shining record of communicating ...

P.S. In addition, most of those command chair posts are a bit light on content to be honest though there are a few gems in among them ...

P.P.S. Anyway, enough of this for now ... back to our regularly scheduled Karl centric programming since it is much more interesting

Edited by Mawai, 12 June 2014 - 08:58 AM.

[Draconis Nagasawa]

Way to much time on your hands if you keep track of all of that stuff

Mawai, on 12 June 2014 - 08:54 AM, said:

Finally, whoever said anything about "to respond to every needy person to full fill (sic) every single question" ... that is both exaggeration and fabrication ...

The truth, in a age/generation of entitlement and gratification.

Now on with the show!

Edited by Crusadiar, 12 June 2014 - 11:49 AM.

[Karl Berg]

Imperius, on 11 June 2014 - 09:20 PM, said:

Karl Berg I have a very important question I would like to ask.

What would be the best PC to build to play MWO at optimal FPS at Very High Settings? I am looking to build a game rig on the cheap and I was wondering what you guys/girls use to play and test the game with. I'm also looking to play it at a resolution no bigger than 1080p. Also I would like it to be able to run those 2-4k texture files you may release in the future.

i7 or i5
RAM Speed (Does it matter to MWO)
SSD or HDD

I have an Nvidia GTX 680 but I think my dual 2.8 Xeons(quad-core) are bottlenecking it's potential. It's a 2008 pc so its time to update I guess.

Hey Imperius!

I saw your post earlier, and was meaning to write down specific specs on my work machine today to give you some potential guidelines. But with the very rocky PTS run today, I completely failed to follow through. I also ended up missing all my meetings too .

Your system looks plenty beefy to be honest. Are those i7 Xeons, or older P4 based cores?

The draw calls made into D3D are very CPU intensive. A good chunk of that is due to the lego-like nature of the mechs; being formed out of dozens of individual components rather than a single character that can be rendered with a single draw call, like in most other games. It's also compounded by the particle system, the terrain system, and the older Scaleform 3 integrated into the engine.

Because of this, you're definitely going to want a very strong CPU. I'd recommend a quad core. I have an older i7 920 at home which does decently well; and I work on an i7 of some newer flavour at work. You'll want a very fast bus to your GPU for efficient main memory to GPU DMA transfers, so PCI x16 or better. If your current Xeons are net burst cores, making the move to i7 based cores would definitely be the most important thing you can do for performance.

Fast main memory, and low latency if you can manage it; especially if we get that high-res 4k texture pack released. I think I run 8/8/8/24 at home, but it's been a few years since I specced the system.

Because we're so CPU limited, most decent GPU's will do a good job, but the more raw fill rate you have, the less the particle system will slow you down. I run a 1GB AMD 5870 at home. Pretty sure I have an nVidia card of some flavour at work, although I'll double check on that for you.

SSD will help load times somewhat, and may also reduce hitching if an asset didn't pre cache correctly. I doubt it would make a difference for general framerate in-game.

Hope that helps!

[Karl Berg]

Mawai, on 10 June 2014 - 05:32 AM, said:

As an example, he made a number of optimistic comments in the latest NGNG podcast regarding the matchmaker. How there are no problems with 4x3 including tonnage matching, that everything looks good. He mentioned that the queue indicator appeared to be nicely evenly out at 25% each even though it isn't doing anything at the moment ... I don't know when he is looking but that isn't my experience. Typically the queue indcator shows 10 to 20% in each of the light and medium classes and 30 to 40% in the heavy/assault classes. This is the natural distribution for player choice of mechs. Will this change when the queues are active ... I hope so ... but I don't think it will change enough. I still think there will be enough folks wanting to play heavies and assaults that substantial queues will form over time. I hope I am wrong but if they don't have a plan in place to deal with it then this is being set up to fail.

Anyway, I am playing wait and see with the revised matchmaker with my fingers crossed ... just as I am for any design element which Russ has commented on ...

I believe he is referring to the queue mix during those times when we attempted to enable 3's. The analytics are biased to some extent, due to the old matchmakers inability to produce games at an acceptable rate, and the extreme queue times all weight queues were experiencing as a result. Even the light queue was starving during this period. Regardless, while 3's were enabled, the queue mixture was pinched very strongly together towards a 25% distribution for each weight class. Even with 3's disabled, the mixture of weight classes aggregated across a single day has shifted quite a lot from where it was before.

The new matchmaker has release valves built in to almost every matching constraint. There are some serious implications resulting from this technical approach; but I've discussed in detail with design what these implications are, how we can control constraint decay, and why these valves are so important. So far, design has been highly supportive. The end goal is for the new matchmaker to be adaptable to almost any player input mixture.

I tried to explain somewhat in the next dev vlog; but didn't manage to express as much as of the algorithms underlying the new system as I wanted. We'll see if you guys still think I'm a good communicator after watching that vlog..

[Alaskan Nobody]

Karl Berg, on 12 June 2014 - 09:46 PM, said:

I tried to explain somewhat in the next dev vlog; but didn't manage to express as much as of the algorithms underlying the new system as I wanted. We'll see if you guys still think I'm a good communicator after watching that vlog..

I can understand that frustration - I spend a lot of time trying and failing to describe ideas and concepts to people.

Don't beat yourself up to much over it if it doesn't come across like you would want

[Chimerahawk]

Mawai, on 11 June 2014 - 11:06 AM, said:

LOL ... well lets see ... I already know what the HM sound is like ... it is annoying ... at least to me ... I also tend to play with the sound turned down due to other people in the house ... having a large number of mechs on the field equipped with a variety of sounds like "Sonic Boom" and others ... can't be any better than a couple of lances of Heavy Metal ... and it is pretty easy to judge based on experience that I would appreciate the ability to toggle this particular sound effect on and off.

The game already has sliders for effect volume, dialog volume, music volume and a checkbox for Frontend Sound in the mechlab ... is it that much of a stretch to request a checkbox to turn on/off warhorns?

I have nothing against the idea or how they choose to implement it but I really don't want to bombarded by tunes and noises when a mech scores a kill.



Well ... I think if you choose to equip one you should be stuck with the look ... but I don't think everyone else in the game should have to listen to it

After testing them on the test server all day, the sound is only really jarring to the person who equipped it. I barely noticed them on other people, as they're not quite as noticable as the heavy metal sound.

So wait, you should get the toggle you want, but other people shouldn't get the just-as-reasonable one they'd like..? That's a bit hypocritical isn't it?

Forgot to mention that I'd like both options to be available.

Edited by Chimerahawk, 12 June 2014 - 10:21 PM.

[zudukai]

Karl Berg, on 12 June 2014 - 09:46 PM, said:

We'll see if you guys still think I'm a good communicator after watching that vlog..

these posts are a clear call of good communication. thank you.

[Heffay]

Karl Berg, on 12 June 2014 - 09:18 PM, said:

Fast main memory, and low latency if you can manage it; especially if we get that high-res 4k texture pack released. I think I run 8/8/8/24 at home, but it's been a few years since I specced the system.

Karl Berg, on 12 June 2014 - 09:18 PM, said:

Fast main memory, and low latency if you can manage it; especially if we get that high-res 4k texture pack released.

Karl Berg, on 12 June 2014 - 09:18 PM, said:

especially if we get that high-res 4k texture pack released.

Karl Berg, on 12 June 2014 - 09:18 PM, said:

high-res 4k texture pack

Woah woah woah.... WOAH! Wait... you have 4k textures for everything???

I've never made a demand from PGI or IGP before, but here is mine: Please please more info???

[Wintersdark]

Heffay, on 13 June 2014 - 05:46 AM, said:

Woah woah woah.... WOAH! Wait... you have 4k textures for everything???

I've never made a demand from PGI or IGP before, but here is mine: Please please more info???

Yeah, Matthew Craig mentioned this in Lordred's OMG MECH PR0N thread - they're considering a High Res Texture pack as "DLC". I'd *love* that; apparently they went down to 1K textures because as patterns and mechs are added the install size was ballooning dramatically.

Personally, I don't care in the slightest how large the game install is, I'd give my left nut for an optional 4k HRTP.


Basically, the base textures are all 4K by default. They're compressed down to 1K (and in a couple cases 2K) to shrink the game's footprint for download and install.

Edited by Wintersdark, 13 June 2014 - 06:59 AM.