AMD Phenom X4 970 Served me Well!

UMASS

Gawd
Joined
Jan 17, 2008
Messages
1,001
Been an AMD fanboy forever. I finally had to make the switch to Intel i7-4790K. My quad-core AMD did a great job & overclocked well (4.1GHz) with a Corsair H50 water cooler (push-pull) I wished I could have purchased an AMD CPU! Dame shame that it has come to this.

A few months ago I was thinking about a 8370 & new MB. It wasn't worth it. I bought my new set-up from Micro Center (40 miles away) So worth it! AMD so needs to step-up & make a decent CPU! Id we lose AMD, then Intel will have their way.
 
I too am disappointed that AMD seems out of touch with the enthusiast community. Both graphics and processors seem like afterthoughts. It's like their heart isn't into it anymore.
 
I completely agree. I hate to buy an FX8320 if the motherboard is going obsolete with ddr4 around the corner. i dont like the APU because they are only quad cores.

I switched to intel in the form of Lenovo Y50 gaming laptop ironically using an AMD FX6300 for a work pc to test and play with windows 8 hypervisor. its doing very well, but not being pushed hard.
 
My first build was a 1045T undervolted and overclocked. Had to drop it when it bottlenecked an OCed 6970. It was still a great CPU though.
 
I still have my Phenom II X6 1075 running at 3.8GHz that I bought late 2010. It's really showing it's age now in games and I'm about ready to upgrade. I am in the same position and thought about an 8350 but it's really not worth it. I don't think I've ever had a CPU last this long. It's been a great 6 core processor.

If AMD had something more competitive I'd probably go that route. A lot of the Intel offerings just seem a bit overpriced or locked into a price and lower too slowly. Oh well.
 
Last edited:
I switched from my 1090T bought when it released. That system lasted me longer than anything I ever had. Started it with a 550 black edition, flashed the bios and kept going with the 1090 when it came out.

It didn't take long to see AMD had nothing with much performance for price value as they used to so I went back to Intel for the first time since the P3. My 7970 will be the next thing to go and it looks like the 300 series may be a pass as well.

I hope they turn it around before much more time passes.
 
AMD desperately needs Zen to be a winner. They've never gone this long without a significant upgrade path with their CPU's. Sucks that all those years of mismanagement bit them in the ass this hard.
 
I miss the massive boost I got from my 955BE back in the day, I had been out of the computer scene since the early 2000's and was still using an old single core socket 939 machine, man that chip felt like an amazing upgrade. I keep buying FX chips now and again just to tinker with but I can't give up the extra performance in games just to float the underdog.
 
I also hold the opinion that Phenom II X4 and X6 series CPUs were the last worthy performance processors released by AMD. Really hope Zen turns out well.
 
As someone who's lost a TON of money in AMD stock value, I am painfully aware of how AMD needs to step up their game :(
 
I also hold the opinion that Phenom II X4 and X6 series CPUs were the last worthy performance processors released by AMD. Really hope Zen turns out well.

I pre-ordered a Bulldozer FX-8150 and ASUS Crosshair V in anticipation of the stellar performance and 8-core awesomeness I was promised. I ended up reverting back to my 1090T because even at stock clocks its performance felt better. The CHV crapped out in short order as well. I ended up selling out and joining the evil empire.

After they had had some time to mature, I revisited the FX lineup with a 6300 briefly, before I realized it was at best on-par with my Sandy Bridge i3 for what I was doing. The only recent AMD chip that I have been completely satisfied with was the A10-7850K, which made a badass HTPC in a form-factor that fit in my entertainment center.

I hope Zen can at least be competitive in the price/performance category, since that is where AMD seems to be focused currently.
 
Guys if I was you, I'd get those 6 core Phenoms on the Ebay market ASAP. Get rid of them before most of us move over to Intel for good.

You'll still get a good price for them for funding your new Intel rig.
 
I don't like the APU because they are only quad cores.
I'm in the same boat. I looked at an apu for a new build but really need 6 cores minimum. The machine in question is a development work station that powers anywhere from 2-3 VMs regularly, and up to 4 less frequently.

I also need support for 3 monitors, which looks sketchy wrt apu motherboards.

I like the idea of an apu build, primarily for added silence, but I'm skipping it over for this build and will see where they're at in a few more years.
 
A few months ago I was thinking about a 8370 & new MB. It wasn't worth it. I bought my new set-up from Micro Center (40 miles away) So worth it! AMD so needs to step-up & make a decent CPU! Id we lose AMD, then Intel will have their way.

And yes it is not good but be a bit more realistic , AMD produces a cpu which is really over 2 years old and has next to no improvements over the original Piledriver in the performance area.

When you want a new cpu get ready for Zen in 2016 Q3 is when we should know a good bit more.

It is weird that people are trying to compare very old cpu architecture with a new one . However if you are gaming with Mantle/DX12 the difference especially on higher batch count games that difference should not be that big..
 
Last edited:
As you can see from my signature I still use quite of bit of AMD for my ancillary builds. NAS, VMWare, etc. I had been a solid AMD user all the way up until Ivy Bridge came out. My quad core Ivy Bridge DESTROYED the 8150 I had. However to this day when someone wants a machine for anything other than gaming I use an AMD processor. I still think they are better value for everyday computing.
 
still using both my K10 based machines and perfectly happy with them... one a dual Opteron 8439SE and the other a quad Opteron 61xx ES (overclocked to 3 GHz). i noticed that AMD just recently surpassed the IPC of the K10 with Carrizo lol :)

Very much looking forward to Zen... and what Jim Keller has up his sleeves... lets just hope the best for AMD.
 
Last edited:
I recently moved to an X58 system with an X5675 from my 980 BE and despite the large gap in benchmarks (cpu), I don't see or feel any difference in general use. I only have an HD7950 so I'm still GPU limited on either system. My heaven and Valley scores were minisculely better on the Xeon. I scored a steal of a deal on this setup otherwise I'd still be using my 980 BE. This hexacore Xeon will last me much longer though.
 
Guys if I was you, I'd get those 6 core Phenoms on the Ebay market ASAP. Get rid of them before most of us move over to Intel for good.

You'll still get a good price for them for funding your new Intel rig.

Good idea! I have an 1100T just sitting here.
 
I too am disappointed that AMD seems out of touch with the enthusiast community. Both graphics and processors seem like afterthoughts. It's like their heart isn't into it anymore.

It's called a lack of money, which leads to a lack of R&D, which leads to subpar releases. From what I can gather online, Intel spends roughly double the money on R&D than ALL the money AMD brings in total per year.
 
If I were to do it over again, I would have got an i7. I don't blame you one bit. Not really enough difference in games for me to make the switch though. And no I don't expect DX12 to even the playing fields either. Mantle didn't make much difference. I will hold onto this 8350 until it is simply too slow to do it anymore.
 
I still think AMD chips are better when running multiple virtual machines. Sadly gaming is in the court of Intel.
 
I am happy with my FX 8350 and FX 8320 but, I am also starting to feel things going a little slower than I would like from time to time. However, that could also be that I am running on the older 990FX chipsets more than a cpu issue. Only the 5820k would be worth it as an upgrade and I would have to spend at least $1800 since I would want to upgrade my work and home PC's at the same time.
 
I hate to break it to you but pretty much any quad core Intel will best AMD's octocores. I mean read the Skylake review. It's been that way since Sandy Bridge.
 
I still think AMD chips are better when running multiple virtual machines. Sadly gaming is in the court of Intel.

Everything I've seen says that outside of very limited multithreading scenarios, Intel's 4 hyperthreaded cores will match or beat AMD's 8 cores in multithreaded tasks, let alone single.
 
I hate to break it to you but pretty much any quad core Intel will best AMD's octocores. I mean read the Skylake review. It's been that way since Sandy Bridge.

Hate to break it to who???? I see no reason to upgrade to a "quad" core when anything less than a 6 core 5820k would be a useless upgrade from an 8 core FX 8350 or 8320 that I have. Besides, let's get real, Intel is not going to provide a 6 core or more processor in their Z170 boards which would mean another upgrade in short order.

Me, I like to have something that lasts a long time which is why I have what I have. If AMD goes out of business or Zen fails completely, I will pick up a 6 or 8 core Intel for both my home and work computers. Also, a 4 core 4790k is not going to multitask with multiple VM's running like my FX 8320 and 8350 do. This is why I am saying that anything less than an 5820k is a waste of money coming from where I am now.
 
Mantle DX12 Vulkan will all make this "discussion" moot. Since the graphics cards are the limiting factor only the amount of cores matter and IPC matters little.

Which means that on 4 cores you do well within 40k-60k batches for 6 cores this will run up to 80K+ batches and the 100K+ batches will be for 8+ cores.

In the end it means that Intel has a bigger market share for desktops but the landscape of gaming means that the cpu changes which are perceived as being needed for high end gaming performance now will fall short.
Not that anyone is going to invest their money in old technology and buy AMD by the boatloads but it does show that when you cripple certain aspects and then remove them it paints a different picture.

Where if you are looking at gaming and videocards and processors .you all know where you can get the most performance. The cpu will be trivial except for the number of cores.
 
Personally, I think that they made a serious error in judgement when they choose not to make an 8 core processor with steamroller and now whatever the latest architecture is called. They would have made money or at least broken even in my opinion.

Instead, what they did is lose mind share since they did not give us what we wanted. Even if Zen is a hit, I think it might be too little, too late for the company as a whole since I am not certain they could get the chip out in quantities that would make them enough money to turn the ship around or right it.
 
Personally, I think that they made a serious error in judgement when they choose not to make an 8 core processor with steamroller and now whatever the latest architecture is called.

Perhaps the clocks had to be reduced on the new process making the IPC improvement a wash.
 
Personally, I think that they made a serious error in judgement when they choose not to make an 8 core processor with steamroller and now whatever the latest architecture is called. They would have made money or at least broken even in my opinion.

Instead, what they did is lose mind share since they did not give us what we wanted. Even if Zen is a hit, I think it might be too little, too late for the company as a whole since I am not certain they could get the chip out in quantities that would make them enough money to turn the ship around or right it.

It wasn't a bad judgement call, they made the right decision.

The original plans was to create new Opterons based on Steamroller and eventually Excavator on smaller process nodes like 20nm and potentially 14nm, but GloFo dropped the ball again and they opted to simply cancel those plans outright. As we saw with 28nm SHP, the process wasn't suited for high clockspeeds, so whilst they could've potentially done some decent server chips with that process, it wouldn't have scaled well at all at higher TDP's, making a client desktop variant pointless.

This is why there were reports in late 2012 saying that AMD had "canned Steamroller and Excavator", which was a half-truth at the time -- those uarchs were getting reworked into what we know now as "SteamrollerB" and even the version of Excavator we see in Carrizo is different than the one that was originally in the cards. They got repurposed for energy efficiency and low-power-oriented APU designs.

AMD poached Jim Keller from Apple in mid-2012 because they already know the Bulldozer uarch was a flop and they wanted a serious new x86 uarch centered on per-core performance rather than gimmicks like CMT. So we'll see in 2016 whether or not it will save AMD from the terrible situation they've found themselves in after years of inner-sabotage and mismanagement, as well as slipups on GloFo's part. Contra-revenue and Intel-OEM tomfoolery notwithstanding.
 
It wasn't a bad judgement call, they made the right decision.

The original plans was to create new Opterons based on Steamroller and eventually Excavator on smaller process nodes like 20nm and potentially 14nm, but GloFo dropped the ball again and they opted to simply cancel those plans outright. As we saw with 28nm SHP, the process wasn't suited for high clockspeeds, so whilst they could've potentially done some decent server chips with that process, it wouldn't have scaled well at all at higher TDP's, making a client desktop variant pointless.

This is why there were reports in late 2012 saying that AMD had "canned Steamroller and Excavator", which was a half-truth at the time -- those uarchs were getting reworked into what we know now as "SteamrollerB" and even the version of Excavator we see in Carrizo is different than the one that was originally in the cards. They got repurposed for energy efficiency and low-power-oriented APU designs.

AMD poached Jim Keller from Apple in mid-2012 because they already know the Bulldozer uarch was a flop and they wanted a serious new x86 uarch centered on per-core performance rather than gimmicks like CMT. So we'll see in 2016 whether or not it will save AMD from the terrible situation they've found themselves in after years of inner-sabotage and mismanagement, as well as slipups on GloFo's part. Contra-revenue and Intel-OEM tomfoolery notwithstanding.

This does make sense. However, they never came right out and announced what was happening but instead, they left us in the dark for a long time. That alone managed to burn up some of whatever good will they had in the community. Also, unfortunately, their APU's, as good as they are, are not really making them any money.

I just do not see a bright future for them even if Zen kicks butt. They need a hugh contract for OEM's and the ability to produce in quantity to be able to start making money.
 
^Didn't Mayweather spent most of the "fight" hugging him and running away from him? I didn't watch it but that's how everyone described it to me, lol.
 
Even if Zen turns out to the PacMan and Intel is still Mayweather, there are many folks who would be more than happy with PacMan.

Yes but there won't be enough of those folks to keep AMD afloat for long if Zen fails to impress.
 
Yes but there won't be enough of those folks to keep AMD afloat for long if Zen fails to impress.

There will always be something wrong with Zen. But will Zen be the best cpu AMD can make and build upon the next several years is the question.

Impressive would be if it came close enough performance wise that is the most people can expect from AMD for AMD which has problems funding the R&D this would be very impressive.

Claiming in any way shape or form that AMD should beat Intel has no basis in this reality.
 
Even if Zen somehow achieves parity with Skylake performance-wise, there will still be people saying "meh, too little too late".
 
Back
Top