Profile of anubis44

my specs

CPU: I7-4790k 4.4(ghz)

GPU: gtx 780

ram: 8g corsair vengeance

mobo: GIGABYTE Z97X-GAMING 7

i play at 1080p all max with v-sync disabled

MY average fps is 67.6

COH 2 is very cpu heavy. i had an a8-3870k at 3(ghz) before

ran coh2 horrible.

Iv'e had very bad experiences with amd hardware. Constant crashing and bad performance. i recommenced getting intel

Thanks for the feedback. Yes, the AMD APUs are terrible for this game, as they lack LVL 3 cache, as are the Intel dual core CPUs, even the ones with hyper threading (Core i3). However, the crashing you're talking about is definitely not the result of an AMD CPU. More likely, you had a driver issue, or even an unstable overclock or an unstable motherboard/GPU. I've only had good experiences with AMD hardware, but then again, I always go for a top-end AMD CPU, never an APU or CPU without LVL3 cache.

The minimum CPU I'd suggest as decent is an AMD Phenom II (not Athlon II) or an Intel Core i5 desktop CPU (the laptop core i5s are really core i3 chips labelled as i5, as they only have two cores + hyperthreading). The AMD FX 6 core or 8 core are better than the Phenom II, and are closer to the i5 than the Phenom II.

Anybody else want to share their system specs/benchmark results?

Hi all. Just wanted to share some up-to-date benchmark results, as many people will only find results that are over a year old at this point when looking at how an AMD FX CPU will perform in this game, and I know how frustrating that can be when you're looking to get new hardware, and you can't find recent info. So...

I have an FX-8350 CPU clocked at 4.5GHz with a very minor voltage bump. Graphics card is a Gigabyte 7950 card (3GB) that I've bios flashed to 1GHz stable. I decided to try to figure out exactly how many cores this game likes with an AMD FX chip, and what effect, if any, using the '1 core per module' setting would have.

I run the game on a 1680x1050 monitor with the following quality settings:

Game Play Resolution: 100%
Image Quality: High
AA: Low (setting this any higher on just about any graphics card will completely cripple the game!)
V-sync: off
Texture Detail: High
Snow Detail: Medium
Physics: Medium (just realized this can be set to high with only a very marginal impact - ie. 2 FPS loss!)

Here are the results of the built-in benchmark - FPS:

With 4 cores enabled:

Min: 23.82
Max: 61.14
Avg: 40.81

With 6 cores enabled:

Min: 32.90
Max: 64.10
Avg: 46.59

With 8 cores enabled:

Min: 32.23
Max: 65.67
Max: 46.94

With 4 cores - 1 per module enabled:

Min: 31.78
Max: 66.07
Avg: 48.30

So, the results I obtained seem to indicate that with the AMD 7950 graphics card I have, at 1680x1050 resolution, and the game graphics settings I used, an AMD FX-6300 or FX-6350 chip clocked at 4.5GHz would give you virtually identical performance to my FX-8350 chip with all 8 cores clocked at 4.5GHz. In fact, the game will run every so slightly faster with only 4 cores enabled, using the 1 core-per-module setting on an 8 core FX CPU at that clock rate! This is quite fascinating, because it makes me wonder why the 4 core FX CPUs do so much worse than the 6 and 8 core FX CPUs? I know the FX-4350 is a two-module chip, but I would have thought there would be much less resource-sharing contention between the chips on the same module than is indicated by my results.

Finally, I just want to say that I think an FX-6300 or FX-6350 chip would be the best bang-for-the-buck chip for someone who is primarily looking to play COH2, and wants to keep their CPU budget as low as possible. I'm quite happy with my FX-8350, and essentially, a 6 core FX chip clocked to the same MHz (4.5GHz) will perform identically well in this game.

Now, if AMD will only release an R9 390(X) graphics card that will double the frame rates of the R9 290(X) cards in this game on my current platform, I will be all over that card!!

Hope some of you find this information helpful!

This is where a lot of the hostility between 1v1/2v2 and 3v3/4v4 players comes from. Does focusing design and balance on 1v1 and 2v2 with some balance drifting up to 3v3 and 4v4 attract more players than focusing balance on 3v3 and 4v4 at the sacrifice of 1v1 and 2v2?

Historically speaking 1v1 and 2v2 attract more players to a RTS game than 3v3 and 4v4. Relic has already stated on the official forums that they realize this and will continue to focus their efforts on 1v1 and 2v2. The problem with 3v3 and 4v4, is that games are hard to cast with so much going on and is hard to watch with the clutter of 8 players' units. Tournaments are hard to run and organize due to having ot have 8 peoples' schedules line up. RNG can be more punishing in team games. If all 4 of you form a concentrated attack of infantry on one side a walking stuka/call-in arty/or Katyusha barrage could wipe the majority of your squads resulting in all of you having massive MP lose.

The game currently has more casual than competitive players and unfortunately the problem with casual players over competitive is that competitive players will stick with a game longer than casuals. Casuals drift more freely from game to game and this can lead to the game base collapsing, competitive players do play other games but they invest a significant amount of more time into this game than casuals. Balancing for casuals might see a small player base increase for a few months, but a year form now, the game will be dead and stale. Balancing for competitive players might hit the player base at first as the casuals get frustrated with the competitive players, but as tournaments and more videos start up the player base will thrive. Casuals will want to try being a little competitive and will play more to get better, competitive players will be practicing and when both of these groups need a break form 1v1 and 2v2 they will go over to 3v3 and 4v4 for some mass unit chaos.

Can 3v3 and 4v4 be more balanced? Yes, but will it being balanced at the cost of 1v1 and 2v2 help the game grow more? No.

I believe it's a mistake to assume that 3 vs. 3 and 4 vs. 4 players are 'casual'. I've been playing COH since 2006, on average several matches per week, and I've never really been interested in 1 vs. 1, and only rarely have I played 2 vs. 2. Am I a 'casual' player? Why don't I play 1 vs. 1 or 2 vs. 2? Because for me, the game stops being a 'game' when there are enough random elements introduced to eliminate the predictable, causality of a small match: in short, it's more like real war, rather than chess, which I find infinitely more stimulating. It's boring to me to realize only 5 minutes into a match that, because you didn't take the 'farm house', you've lost, and it's GG. Utter stupidity as far as I'm concerned. In a 4 vs. 4, if you are having trouble taking the 'farm house', you can ask an ally to provide reinforcements, while still leaving 2 other teammates to continue with other objectives, and possibly reverse a potentially serious setback without simply having to 'GG' and start again. Or you could find that enemy has allocated 3 players worth of troops to take an objective rather than only one player's worth of troops -- the random element is much greater with more players. In a 1 vs. 1, you know EXACTLY how much force your opponent can possibly have, and the only variable will be where they are. Boring IMHO. Kind of like being able to count cards. I crave the sheer unpredictability of war. It's like playing chess, but your opponent might have 3 bishops instead of 2, or 2 queens and no knights, etc.

In addition, as I mentioned in an earlier reply, I find the chatter of comrades over the headset adds a level of verisimilitude that's missing if you're by yourself, or only communicating with one other ally. The chaos of trying to get your colleagues' attention during a crucial battle event is something that's utterly missing in a 1 vs. 1. I'm sure I'm not the only player who craves that kind of auditory battle-atmosphere. In fact, I think a future version of COH should require players to connect their headsets into the game engine itself, so that if/when one of the 'radio intercept' abilities is activated, you can actually listen in to the chatter of opposing team for a few seconds without them knowing. Yeah, I know there are potential issues with this (people refusing to patch in, and using skpye instead, players not speaking the same language, etc.) You could require the players to patch in to the game with headsets before the match starts, and use voice-pattern recognition to force them to really do it or the game will not start/or it will stop in mid-match, but I digress.

for such an "epiphany" its not really well thought out, presented and tbh im not sure it even can be qualified as an epiphany.

my opinion on the topic non withstanding. he has every right to make this thread reguardless of your/my/bob joe's opinions about it. if a moderator thinks a thread is breaking of the rules he will deal with it. posting "off topic" in the thread isn't the proper way to handle this we have a report button for that. also someones trash may very well be someone else's treasure.

if you think a thread is breaking the forum rules please use the report button. if you dislike the contents for whatever reason. there is the option to ignore it or explain why you disagree.

TL;DR
Show Spoiler

It's his epiphany, and he's entitled to it.

Thanks all for your replies.

I decided to spend a bit more money on this upgrade. With Bill Me Later option on newegg.com I could spend $500 bucks and get it paid off in 6 months with no problem.

So what kind of game settings will I get with these build?

I5 4430 3.0 MHz CPU
Radeon HD 7870 2GB GZH edition
Gigabyte B85 HD3 motherboard
Windows 7
G. Skill Ripjaws 8GB RAM DDR3 1600mhz ( I can always upgrade it to 16GB)


Also, don't forget that when you get new CPU you need a new mobo. When you get new mobo you need new Windows. So it's impossible to just simply upgrade your CPU. You need to build a new rig pretty much from the bottom and up.

That i5 CPU and the 7870 GPU represent the best possble bang-for-the buck COH2 system components. Well done. If I may be so bold, allow me to recommend the Gigabyte Windforce 3 7870, which is factory overclocked out of the box at 1100MHz on the core, and represents excellent value in a 7870 card. Spending any more money on either the CPU or the GPU will only provide you with very incremental improvements in FPS which likely aren't worth the money, especially on the CPU side.

However, if you really want more performance, on the GPU side, the next really noticeable step up (I'll assume you're using a 1920x1080 monitor) from the 7870 would be either a 7970 (now called the R9 280X) for about $300, and the now-just-released R9 290 (non-'X' version) for $399. The difference in FPS can be seen in this benchmark:

http://www.anandtech.com/show/7481/the-amd-radeon-r9-290-review/5

Looking at the 1920x1080 benchmark, you can substitute the 7870 for the '7950b' number, as they're going to be virtually the same (give or take 1 or 2 FPS). So the i5+7870 will be giving you just under 40FPS in this quite demanding benchmark, which represents a worst-case scenario in COH2. For another ~$100 or so over the 7870, the R9 280X will merely buy you about another ~7 or 8 FPS, putting you at ~45FPS -- not the best way to spend $100 in my opinion, although you will also get significantly more performance in other games with this upgrade. Finally, by spending ~$200 more than the 7870, you can get an R9 290 card which will give you ~20 extra FPS in this benchmark, putting you at about 60FPS. If you really want to spend this extra $200 for the extra headroom it will buy you, however, I very strongly recommend that you wait for the custom cooled versions of the R9 290 that will certainly be coming out (probably by end of November) from Gigabyte, Asus, MSI, Sapphire, Club 3D, etc., and will certainly be much quieter than the just-released reference R9 290 and likely overclocked, too. These cards will likely represent the pinnacle of COH2 performance, along with the R9 290X custom cards (when they are also released).

Finally, remember that COH2 DOES NOT CURRENTLY SUPPORT MULTI-GPU, so SLI and Crossfire setups will NOT provide any additional performance over a single-card setup of the same type. This means it makes more sense to buy a single $400 R9 290 than two $200 GTX660s or two $200 7870s. This MAY change in future, but I wouldn't risk betting on it for this game, at this point.

Good luck!