AMD Takes The CPU Performance Crown For The First Time in Almost a Decade.

[leipero]

Oh right, so my list is irrelevant because it disproves OP? So if I make a thread which says "Intel is better than AMD", then only post 2 benchmarks, both of which I have carefully cherry picked in order to avoid posting anything where AMD beats Intel, then anything that shows AMD being better is "irrelevant"? The benchmarks I posted from that shady Russian site (same as OP used) are very relevant to this thread, because they exposes how biased OP is. You can't say it is irrelevant just because it proves that OP is flat out wrong.

At this point, after reading all your posts (even the last one i didn't quoted), I'm starting to wonder if you (and some others) really read OP and title of the topic. Anyways, have fun.

[Andrew0472]

The PC upgrade market is going to come to a crawl soon. You simply won't need to upgrade. Well, maybe for the odd game here and there that is actually made from the ground up for PC (like Arma, and, well I can't even think of any others).

 

The race is pretty much coming to a conclusion. Clock speeds are not really getting much higher, it's not about shrinking dies and reducing power consumption. Which is good for laptops and tablets.

 

AMD have secured their nest egg, Intel now need to find one. Not quite sure what will happen to Nvidia, but maybe they should stop letting brain dead monkeys price their products (I'm talking about Titan of course, ridiculous price).

intel are getting roughly the same performance if not more out of their chips which run cooler and consume less power but amd are settled? that doesn't make sense 

[eshbop]

At this point, after reading all your posts (even the last one i didn't quoted), I'm starting to wonder if you (and some others) really read OP and title of the topic. Anyways, have fun.

OP makes a big point of AMD 8cores for "future-proofing".

 

Too me, there's no sense in future-proofing (primarily mobo, CPU, GPU) because I want something that works well now, not might be good in the future. Also, in this community these components have relatively frequent upgrade cycles which means when the future arrives you can make an informed choice.

 

 

And again, with a "good enough" CPU the GPU will be the thing holding you back. Also, the AMD platform has a bit of catching up to do before it's get the features of Haswell.

[soulreaper]

Well intel has more "ppc" so all intel needs to do is roll out more iterations of the hexacore and make an octacores. But it will probably cost an arm and a leg as opposed to AMD.

[ALXAndy]

intel are getting roughly the same performance if not more out of their chips which run cooler and consume less power but amd are settled? that doesn't make sense 

 

Intel chips run cooler? funny that one.

 

Intel I5 4670k TJMAX (or thermal junction if you didn't know what it means) = 105 c.

 

AMD FX 8320 TXMAX = 68c.

 

Haswell chips run about as cool as the surface of the sun.

[LAwLz]

At this point, after reading all your posts (even the last one i didn't quoted), I'm starting to wonder if you (and some others) really read OP and title of the topic. Anyways, have fun.

Yes I did, it says "AMD Takes The CPU Performance Crown For The First Time In Almost a Decade". What is your point? That I am not allowed to disprove OP? Again, I am just pointing out that OP is lying by cherry picking benchmarks and trying to deceive people inot thinking AMD is better than they really are right now. Again, if I am missing your point then please explain what your point is.

 

 

OP makes a big point of AMD 8cores for "future-proofing".

 

Too me, there's no sense in future-proofing (primarily mobo, CPU, GPU) because I want something that works well now, not might be good in the future. Also, in this community these components have relatively frequent upgrade cycles which means when the future arrives you can make an informed choice.

 

And again, with a "good enough" CPU the GPU will be the thing holding you back. Also, the AMD platform has a bit of catching up to do before it's get the features of Haswell.

Well it's good to see that we agree on something, that future proofing things like CPUs makes no sense.

 

 

Intel chips run cooler? funny that one.

 

Intel I5 4670k TJMAX (or thermal junction if you didn't know what it means) = 105 c.

 

AMD FX 8320 TXMAX = 68c.

 

Haswell chips run about as cool as the surface of the sun.

Do you even know what tj max means? It is completely irrelevant to how much heat a chip produces. The t case temp for the i5-4670K you cited is 72.72 degrees Celsius by the way but again that's not relevant to how much heat a chip produces, just how much heat they are able to handle. I couldn't find what the 8320 is rated for on AMD's site so I'd like a source on where you found that.

If you want a better indicator of how much a chip produces then you should look at the thermal design power (TDP). It refers to the maximum amount of power your cooler has to take care of at any moment. If you got a TDP of 100W, then your cooler should be able to handle 100W of heat output. The i5-3670K has a TDP of 84W, so your cooler has to be able to handle 84W of heat output. The 8320 has a TDP of 125W, so your cooler has to be able to handle 125W of heat output.

Which one do you think produces the most heat?

[Andrew0472]

Intel chips run cooler? funny that one.

 

Intel I5 4670k TJMAX (or thermal junction if you didn't know what it means) = 105 c.

 

AMD FX 8320 TXMAX = 68c.

 

Haswell chips run about as cool as the surface of the sun.

not really sure what you are trying to prove showing that intel chips can safely run hotter than amd..

[Kuzma]

It's funny, you know?  When Intel releases a CPU with marginally better performance and less power consumption, everyone goes "Intel! We don't care about power consumption, just make it more powerful!"  Then AMD goes and releases a powerful CPU with no cares about TDP and everyone's like "What's with that TDP?  We don't want anything like that!"

exactly Glenwing

&

@LAwLz

Where do I begin to tell you exactly how wrong you are, you said the OP cherry picked his benchmarks; of course he cherry picked them, he picked the ones that have been programmed correctly - he picked the ones that have been optimised to take advantage of additional threads regardless of AMD or Intel. He isn't looking at single core games he is looking at the future, all the aspects yourself and others have mentioned "heat" "power consumption" etc. are completely irrelevant because all the OP is saying is that the AMD processors won in this benchmark for the price, end of discussion. As Glenwing said above you're the same kind of person to complain about Intel not improving performance much but lowering TDP and then complain about AMDs CPUs outperforming at high TDPs all I see there is plain simple pure hypocrisy .

[Lays]

Tbh.. the 9590 is at 5 ghz or w/e.. if you oc a 3930k to that, it walks all over it lol, im not really seeing how AMD wins this, considering they're overclocked chips.. To compare these you would need to OC the other chips from intel.

[Kuzma]

Tbh.. the 9590 is at 5 ghz or w/e.. if you oc a 3930k to that, it walks all over it lol, im not really seeing how AMD wins this, considering they're overclocked chips.. To compare these you would need to OC the other chips from intel.

When Intel release a specially binned 3930k that runs at 5 Ghz we can talk , it's like saying the 3960x will walk over the 3970x when it's overclocked because that's obvious: we're talking stock clocks here.

[Kuzma]

Yes I did, it says "AMD Takes The CPU Performance Crown For The First Time In Almost a Decade". What is your point? That I am not allowed to disprove OP? Again, I am just pointing out that OP is lying by cherry picking benchmarks and trying to deceive people inot thinking AMD is better than they really are right now. Again, if I am missing your point then please explain what your point is.

 

 

Well it's good to see that we agree on something, that future proofing things like CPUs makes no sense.

 

 

Do you even know what tj max means? It is completely irrelevant to how much heat a chip produces. The t case temp for the i5-4670K you cited is 72.72 degrees Celsius by the way but again that's not relevant to how much heat a chip produces, just how much heat they are able to handle. I couldn't find what the 8320 is rated for on AMD's site so I'd like a source on where you found that.

If you want a better indicator of how much a chip produces then you should look at the thermal design power (TDP). It refers to the maximum amount of power your cooler has to take care of at any moment. If you got a TDP of 100W, then your cooler should be able to handle 100W of heat output. The i5-3670K has a TDP of 84W, so your cooler has to be able to handle 84W of heat output. The 8320 has a TDP of 125W, so your cooler has to be able to handle 125W of heat output.

Which one do you think produces the most heat?

The thing is that the growth in heat as you overclock intel processors ivybridge or older is exponential whereas with AMD chips it's fairly linear - you reach a thermal limit of Intel CPUs faster due to the manufacturing process - this is why AMD CPUs get further on Liquid Nitrogen and Liquid Helium, because they handle heat a lot better even though they may output a lot of it - AMD have squeezed so much out of their manufacturing process once they have a die shrink they will have fairly big performance increases because they will continuously improve architecture to improve performance as opposed to Intel that rely on die shrinks with their ticktock system

[Lays]

When Intel release a specially binned 3930k that runs at 5 Ghz we can talk , it's like saying the 3960x will walk over the 3970x when it's overclocked because that's obvious: we're talking stock clocks here.

 

 

the 9590 isn't stock clocked though, it's a factory oc'd 8350.. You can't compare an overclocked chip to stock chips...

[Kuzma]

the 9590 isn't stock clocked though, it's a factory oc'd 8350.. You can't compare an overclocked chip to stock chips...

A 3970x is a factory OC'd 3960x and a 2700k is a factory OC'd 2600k that's why I mentioned it - also the 9590 uses less power than an 8350 does at 5 Ghz. It is still the stock speeds if you like it or not so no personal OC'ing occurred.

[Humbug]

the 9590 isn't stock clocked though, it's a factory oc'd 8350.. You can't compare an overclocked chip to stock chips...

Stock clock means the default clockspeed for a particular CPU, whatever AMD or Intel chooses it to be.

 

Overclocking is when the buyer himself increases the clockspeed. This is outside the operating range where Intel/AMD guarantees it to work properly. With the 9590 AMD guarantees it to work at this high clockspeed.

 

Also you can't call the 9590 a factory overclock because this implies there is no other difference between the chips, whereas in reality they are binned differently to the 8350. Which is why it's it's difficult to get a 8350 fully stable at 5Ghz and reasonable voltage; you need some luck.

[Spannerhands]

the 9590 isn't stock clocked though, it's a factory oc'd 8350.. You can't compare an overclocked chip to stock chips...

 

I am sorry but i am really getting board of this comment....

 

Wiki "Overclocking is the process of making a computer or component operate faster than the clock frequency specified by the manufacturer by modifying system parameters (hence the name "overclocking"). Operating voltages may also be changed (increased), which can increase the speed at which operation remains stable. Most overclocking techniques increase power consumption, generating more heat, which must be dispersed if the chip is to remain operational."

 

The FX9590 is 4.7Ghz Base Clock with a 5.Ghz Turbo on one to two cores... It does not even run 5.0Ghz on all cores... So in benchmarks with games like BF4 is only running 4.7Ghz because its using more than 2 cores...

[Kuzma]

I am sorry but i am really getting board of this comment....

 

Wiki "Overclocking is the process of making a computer or component operate faster than the clock frequency specified by the manufacturer by modifying system parameters (hence the name "overclocking"). Operating voltages may also be changed (increased), which can increase the speed at which operation remains stable. Most overclocking techniques increase power consumption, generating more heat, which must be dispersed if the chip is to remain operational."

 

The FX9590 is 4.7Ghz Base Clock with a 5.Ghz Turbo on one to two cores... It does not even run 5.0Ghz on all cores... So in benchmarks with games like BF4 is only running 4.7Ghz because its using more than 2 cores...

You've just proven the point he's saying the 9590 is invalid because its oc'ed but that is the frequency specified by the manufacturer.

[Lays]

I am sorry but i am really getting board of this comment....

 

Wiki "Overclocking is the process of making a computer or component operate faster than the clock frequency specified by the manufacturer by modifying system parameters (hence the name "overclocking"). Operating voltages may also be changed (increased), which can increase the speed at which operation remains stable. Most overclocking techniques increase power consumption, generating more heat, which must be dispersed if the chip is to remain operational."

 

The FX9590 is 4.7Ghz Base Clock with a 5.Ghz Turbo on one to two cores... It does not even run 5.0Ghz on all cores... So in benchmarks with games like BF4 is only running 4.7Ghz because its using more than 2 cores...

 

>corrects someone

 

>doesn't spell bored right

 

I'm saying in theory it is an 8350 with an OC on it, can you push a 9590 further than it's default clock? if you want to melt your board sure, doesn't make sense to me comparing 2 things that are completely different

[LAwLz]

Where do I begin to tell you exactly how wrong you are, you said the OP cherry picked his benchmarks; of course he cherry picked them, he picked the ones that have been programmed correctly - he picked the ones that have been optimised to take advantage of additional threads regardless of AMD or Intel. He isn't looking at single core games he is looking at the future, all the aspects yourself and others have mentioned "heat" "power consumption" etc. are completely irrelevant because all the OP is saying is that the AMD processors won in this benchmark for the price, end of discussion. As Glenwing said above you're the same kind of person to complain about Intel not improving performance much but lowering TDP and then complain about AMDs CPUs outperforming at high TDPs all I see there is plain simple pure hypocrisy .

Ahh yes the "he cherry picked the ones who are correctly programmed". I assume you by "correctly programmed" means "the ones programmed to show the results I want to see", correct? So FarCry 3 is "correctly programmed" but Far Cry 3 Blood Dragon (you know, based on Far Cry 3) isn't? That makes no sense.

When did I mention heat or power consumption? I was strictly taking about performance, and how full of shit OP is for cherry picking the only 2 benchmarks where AMD wins, and ignoring the other 33 where they don't win. I have already addressed the whole "you're the person who whines about lower TDP but no performance improvement but whines about AMD's high TDP" before, it's like 2 pages back. You're just making a pathetic attempt at a strawman argument right now.

 

 

The thing is that the growth in heat as you overclock intel processors ivybridge or older is exponential whereas with AMD chips it's fairly linear - you reach a thermal limit of Intel CPUs faster due to the manufacturing process - this is why AMD CPUs get further on Liquid Nitrogen and Liquid Helium, because they handle heat a lot better even though they may output a lot of it - AMD have squeezed so much out of their manufacturing process once they have a die shrink they will have fairly big performance increases because they will continuously improve architecture to improve performance as opposed to Intel that rely on die shrinks with their ticktock system

No, it's very exponential on AMD chips as well. Not as much as Ivy and Haswell, but it is by no means linear. Not sure what you mean by "you reach a thermal limit of Intel CPUs faster due to the manufacturing process". You reach the thermal limit on Intel chips faster because Intel fucked up, and continues to fuck up on purpose, the connection between the die and the IHS. You are also agreeing with me. I replied to someone saying that Intel chips produces more heat because their TJ max is higher (which shows that person has no idea what TJ max is and should not be trusted) which is simply not true.

How good something overclocks with LN2 has nothing to do with how well they handle high temperatures though, it's exactly the other way around. It's how well they handle extremely low temperatures, and how much voltage they can handle. Even if the t case max was rated for like 150 degrees it might still overclock worse than a chip with a t case max rating of 50 if you are using LN2, because neither of those chips will be even close to 0 degrees while being cooled by LN2.

[Spannerhands]

>corrects someone

>doesn't spell bored right

I'm saying in theory it is an 8350 with an OC on it, can you push a 9590 further than it's default clock? if you want to melt your board sure, doesn't make sense to me comparing 2 things that are completely different

Sorry dyslexic big time, and can't see my errors...

My FX9590 is stable at 5.1 on all cores with a low voltage... Could overclock higher but have issues with ram can get 5.4 but ram causes BSOD.

[Spannerhands]

You've just proven the point he's saying the 9590 is invalid because its oc'ed but that is the frequency specified by the manufacturer.

 

All manufactures bin there CPU's & GPU's and set the "specified frequency" the second that AMD brings out the FX9590 everyone is OMG its just an overclocked FX8350 like its something that's not been done before, its been done by every company for as long as i can remember, just don't understand why everyone makes a big deal abut it?

[Kuzma]

Ahh yes the "he cherry picked the ones who are correctly programmed". I assume you by "correctly programmed" means "the ones programmed to show the results I want to see", correct? So FarCry 3 is "correctly programmed" but Far Cry 3 Blood Dragon (you know, based on Far Cry 3) isn't? That makes no sense.

When did I mention heat or power consumption? I was strictly taking about performance, and how full of shit OP is for cherry picking the only 2 benchmarks where AMD wins, and ignoring the other 33 where they don't win. I have already addressed the whole "you're the person who whines about lower TDP but no performance improvement but whines about AMD's high TDP" before, it's like 2 pages back. You're just making a pathetic attempt at a strawman argument right now.

No, it's very exponential on AMD chips as well. Not as much as Ivy and Haswell, but it is by no means linear. Not sure what you mean by "you reach a thermal limit of Intel CPUs faster due to the manufacturing process". You reach the thermal limit on Intel chips faster because Intel fucked up, and continues to fuck up on purpose, the connection between the die and the IHS. You are also agreeing with me. I replied to someone saying that Intel chips produces more heat because their TJ max is higher (which shows that person has no idea what TJ max is and should not be trusted) which is simply not true.

How good something overclocks with LN2 has nothing to do with how well they handle high temperatures though, it's exactly the other way around. It's how well they handle extremely low temperatures, and how much voltage they can handle. Even if the t case max was rated for like 150 degrees it might still overclock worse than a chip with a t case max rating of 50 if you are using LN2, because neither of those chips will be even close to 0 degrees while being cooled by LN2.

It has nothing to do with Intel fucking up look at Ivybridge-e it uses solder but doesn't OC as well as Sandybridge-e does it? Once you shrink the transistor length to a certain size it becomes harder to dissipate that heat to the heat spreader and therefore cool it down.

By thermal limit (not talking TJ Max, if I ment Thermal Junction Max I would have said Thermal Junction max thank you) I meant the brick wall in which Intel CPUs struggle to continue to OC even on ln2 / lhe because eventually they hit a brick wall much lower than AMDs look at AMDs 8.2 and 8 Ghz world records and then look at Intel's records and Far Cry 3 and Far Cry 3 Blood Dragon are very different games, Blood Dragon was a rushed game in comparison to the former and they most likely wouldn't have bothered to multi-thread the game. And my "programmed correctly" I mean optimised for running on 8 cores. The 33 benchmarks you pulled are irrelevant since they're all old games and/or games that are only optimised for single threads. BF4 is obviously going to perform better on an AMD cpu because of the optimisations for more threads; when we get Watch Dogs benchmarks in I'm sure it'll be exactly the same because developers (such as myself) are moving towards parallelism because there are so many benefits.

[LAwLz]

It has nothing to do with Intel fucking up look at Ivybridge-e it uses solder but doesn't OC as well as Sandybridge-e does it?

It seems like Ivy Bridge E overclocks slightly better than Ivy Bridge, and as far as I can tell it simply becomes unstable (take Anand's review where he says "Moving to higher voltages didn't help, so I had to back down on frequency" as an example). In most reviews I've seen, such as HardwareCanucks, the 4830K is significantly cooler than the i7-4770K (sadly the only LGA 1155 they got on their test) even though the 4770K uses less power and has a lower TDP.

Didn't notice this before but damn it's hard to find temperature tests for CPUs. The HardwareCanucks was the only thing I found for temperature tests on the 4830K.

 

 

By thermal limit (not talking TJ Max, if I ment Thermal Junction Max I would have said Thermal Junction max thank you) I meant the brick wall in which Intel CPUs struggle to continue to OC even on ln2 / lhe because eventually they hit a brick wall much lower than AMDs look at AMDs 8.2 and 8 Ghz world records and then look at Intel's records and Far Cry 3 and Far Cry 3 Blood Dragon are very different games, Blood Dragon was a rushed game in comparison to the former and they most likely wouldn't have bothered to multi-thread the game. And my "programmed correctly" I mean optimised for running on 8 cores. The 33 benchmarks you pulled are irrelevant since they're all old games and/or games that are only optimised for single threads. BF4 is obviously going to perform better on an AMD cpu because of the optimisations for more threads; when we get Watch Dogs benchmarks in I'm sure it'll be exactly the same because developers (such as myself) are moving towards parallelism because there are so many benefits.

Intel's record? About 7GHz isn't that bad. You're making it sound like nobody has ever managed more than like 5GHz on Intel chips, which is far from true (as you can see in the links I posted, and that's just Haswell).

Far Cry 3 and Far Cry 3 Blood Dragon are very, very similar. Blood Dragon is basically a mod for Far Cry 3. The code should be very similar. Just because a game can use more than 4 cores does not mean it benefits from it. Hell WoW has a ton of threads which can run on a ton of cores, but only ~2 of them are actually important and the rest are very very light which might as well run on 1 core. No, not all games on my list are old, and you can't just say "well only these 2 games are relevant and the rest are not I said so". If you're going to use games for benchmarking then you can't just pick the games that show the results you want. That's like me saying "well Intel has the best CPUs because they win in Cinebench, the ones AMD wins in are poorly coded lol". If you're going to use games as benchmarks then you need to take into consideration a lot of them, you know games people actually play. You can't just say "I am going to use games as benchmarks but I will completely ignore the majority of games people play because my beloved AMD loses in them".

The more benchmarks you include, the fairer the comparison because it gives a broader picture. OP is a lying swine who is trying to deceive people, and you're doing everything a disservice by encouraging such behavior. If you're going to tell the truth, then tell the whole truth. Don't hide parts of the truth you don't like.

 

You're a game developer? You're 15... I highly doubt that you've actually made any advanced game.

 

My problem with your and OP's post is that you are basing your arguments on a lot of "if" and future predictions. We have no idea if future games will be able to utilize 8 cores well. You have to remember that most processing used in a game runs in a single chain of dependencies, which is why games usually don't take advantage of extra cores. For example you can't calculate the AI before certain things like physics are done being computed (because if you did you might end up having NPCs running through objects, shooting at things that are already dead and so on). There aren't really any elegant solutions around the problem of dependencies in modern 3D games. Not sure if you listen to Anand's podcasts but they talked about an Android benchmark which calculates chess movies. That's a good example of a game which could utilize a ton of cores very well, because you could assign one core to each chess piece and they can be calculated in parallel. You can't do that with NPCs in a 3D game though because they depend on a lot of other things (so certain tasks has to finish before the computer can even start calculating the AI).

[Kuzma]

It seems like Ivy Bridge E overclocks slightly better than Ivy Bridge, and as far as I can tell it simply becomes unstable (take Anand's review where he says "Moving to higher voltages didn't help, so I had to back down on frequency" as an example). In most reviews I've seen, such as HardwareCanucks, the 4830K is significantly cooler than the i7-4770K (sadly the only LGA 1155 they got on their test) even though the 4770K uses less power and has a lower TDP.

Didn't notice this before but damn it's hard to find temperature tests for CPUs. The HardwareCanucks was the only thing I found for temperature tests on the 4830K.

Intel's record? About 7GHz isn't that bad. You're making it sound like nobody has ever managed more than like 5GHz on Intel chips, which is far from true (as you can see in the links I posted, and that's just Haswell).

Far Cry 3 and Far Cry 3 Blood Dragon are very, very similar. Blood Dragon is basically a mod for Far Cry 3. The code should be very similar. Just because a game can use more than 4 cores does not mean it benefits from it. Hell WoW has a ton of threads which can run on a ton of cores, but only ~2 of them are actually important and the rest are very very light which might as well run on 1 core. No, not all games on my list are old, and you can't just say "well only these 2 games are relevant and the rest are not I said so". If you're going to use games for benchmarking then you can't just pick the games that show the results you want. That's like me saying "well Intel has the best CPUs because they win in Cinebench, the ones AMD wins in are poorly coded lol". If you're going to use games as benchmarks then you need to take into consideration a lot of them, you know games people actually play. You can't just say "I am going to use games as benchmarks but I will completely ignore the majority of games people play because my beloved AMD loses in them".

The more benchmarks you include, the fairer the comparison because it gives a broader picture. OP is a lying swine who is trying to deceive people, and you're doing everything a disservice by encouraging such behavior. If you're going to tell the truth, then tell the whole truth. Don't hide parts of the truth you don't like.

You're a game developer? You're 15... I highly doubt that you've actually made any advanced game.

My problem with your and OP's post is that you are basing your arguments on a lot of "if" and future predictions. We have no idea if future games will be able to utilize 8 cores well. You have to remember that most processing used in a game runs in a single chain of dependencies, which is why games usually don't take advantage of extra cores. For example you can't calculate the AI before certain things like physics are done being computed (because if you did you might end up having NPCs running through objects, shooting at things that are already dead and so on). There aren't really any elegant solutions around the problem of dependencies in modern 3D games. Not sure if you listen to Anand's podcasts but they talked about an Android benchmark which calculates chess movies. That's a good example of a game which could utilize a ton of cores very well, because you could assign one core to each chess piece and they can be calculated in parallel. You can't do that with NPCs in a 3D game though because they depend on a lot of other things (so certain tasks has to finish before the computer can even start calculating the AI).

I agree that Ivybridge-e does overclock a bit better than Ivybridge itself but the instability even at higher voltages is because as I said, once you shrink the transistor size by too much it becomes hard to dissipate the heat to the IHS and therefore to any heatsink/waterblock which is why delidded Ivybridge/Haswell CPUs with cooling applied directly to the die overclock quite nicely - because you're no longer relying on the IHS to dissipate heat.

The fact that you're saying Blood Dragon is like a mod for Far Cry 3 reinforces my point, look at all the Skyrim mods that are terribly optimised for multi-threading and trust me I know how difficult and annoying it is to try and multi-thread everything (including AI) and especially Physics which is why I have opted for use of CryEngine 3 for my team, the engine is beautifully optimised pretty much right out of the box and although the Farcry 3 engine is a derivative of CryEngine 3, it derived from CryEngine 3 in relatively early stages and therefore a lot will be similar but a lot will also be different.

In regards to my age thank you for mentioning it; next month is my 16th birthday and will also be my 10 year anniversiary of programming in some way, sure I may be young but while most kids were playing with their action figures etc. I was learning to code, I learned HTML (doesn't really count since it's so easy but meh) , PHP , MySQL , MSSQL , Visual Basic and C++ by the age of 10 at which point I began to study game engines an proceeded to learn C# because that was the original reason I learned any languages - to make games; now if you wanna continue to say "You're 15 you couldn't possibly have worked on anything advanced" fine , I'd show you past work but I lost everything in a HDD failure 2 years ago, but I don't think the 4 years I've spent working on projects is anything to just ignore & neither is my current game devlopment, we're on a short hiatus currently but as soon as we get our first alpha released you'll see .