This was the best review site around for years.
Be warned that it has become quite the opposite in the last 2 years.
Anand Shimpli hardly does any reviews anymore, but when HE does a review on anything it is worth reading.
His newest reviewers can't spell, and simply spew out his talking points and quite often are just stupid idiots, that are Biased by advertisers, and just a general lack of knowledge.
Here's a prime example of the idiots he left in charge.
http://www.anandtech.com/guides/showdoc.aspx?i=3483
At least he still goes to the shows and has Intel and many other top tier manfacturues kissing his ass.
If you don't see his name as the reviewer, give it a pass.
If Raptors had 32MB's of Cache I would buy 3 of them and never look back.
His statements are false, and he doesn't know the difference from his ass and a hole in the ground.
Too bad though. I would jump on the 32MB VelociRaptors quickly, as they would be the fastest streamers ever made.
Anandtech's Review Site
Re: Anandtech's Review Site
The main target audience for the Velociraptors are people using them in RAID arrays, with Areca SATA controllers that have integrated cache & bbu, and occasionally LSI/Adaptec SAS/SATA raid controllers with the same. In this situation the drives offer unmatched performance unless you're willing to spend a bundle on some legacy u320 gear or modern SAS stuff. Still, if you were to use a Velociraptor as a single drive it's still a very quick drive and the 16mb cache doesn't hurt as bad when you're at 10k rpm spindle speeds with lower platter densities.
As for Anandtech, I'd agree has its share of incorrect specs from some of these newer reviewers but I don't think this is an attempt at spreading incorrect info. Rather these guys seem to be lacking a proper editorial review for many of their articles, treating the site more like a 'blog' than a 'review' site. Seeme to me that the real reason for this is bleedover from the Dailytech news site, with quite a few of the article authors beginning there and migrating over to the main anandtech site.
Still Anandtech & xbitlabs are miles above Tom's and the many 'overclocker'/gamer focused tech sites out there. Both have been rather free of the sort of hijinx that TomsHardware became famous for ('paid' review allegations, firing staff writers & wholesale renaming the authors for their articles, allegedly being paid by AMD to fan the flames of the fanboys, etc).
Another site I've long read, Arstechnica is definitely leaning even MORE towards the 'blog' style, and has far more innacuracies than I see at anandtech, and yet it's still a consistently better read than most places. Just sad that they've lost their deep-tech focus (whitepapers and the like).
Unfortunately I don't see ANY website that consistantly offers good coverage of server/workstation and higher end gear that isn't 'overclocker'/enthusiast/gamer focused. There are some industry specific sites that go beyond simple benchmarks using games, encoding & the occasional cinebench run but they sadly don't do regular coverage (thinking mostly of the professional graphics arena here). Looking at 'computer' tech magazines most follow the PCMag formula of endlessly covering the same entry level & novice discussions targeting business sales & IT departments. Focusing on audio & graphics specific magazines reveals even LESS technical info, and computer technology reviews are often just paid product placements for specific aftermarket OEMs and the like.
As for Anandtech, I'd agree has its share of incorrect specs from some of these newer reviewers but I don't think this is an attempt at spreading incorrect info. Rather these guys seem to be lacking a proper editorial review for many of their articles, treating the site more like a 'blog' than a 'review' site. Seeme to me that the real reason for this is bleedover from the Dailytech news site, with quite a few of the article authors beginning there and migrating over to the main anandtech site.
Still Anandtech & xbitlabs are miles above Tom's and the many 'overclocker'/gamer focused tech sites out there. Both have been rather free of the sort of hijinx that TomsHardware became famous for ('paid' review allegations, firing staff writers & wholesale renaming the authors for their articles, allegedly being paid by AMD to fan the flames of the fanboys, etc).
Another site I've long read, Arstechnica is definitely leaning even MORE towards the 'blog' style, and has far more innacuracies than I see at anandtech, and yet it's still a consistently better read than most places. Just sad that they've lost their deep-tech focus (whitepapers and the like).
Unfortunately I don't see ANY website that consistantly offers good coverage of server/workstation and higher end gear that isn't 'overclocker'/enthusiast/gamer focused. There are some industry specific sites that go beyond simple benchmarks using games, encoding & the occasional cinebench run but they sadly don't do regular coverage (thinking mostly of the professional graphics arena here). Looking at 'computer' tech magazines most follow the PCMag formula of endlessly covering the same entry level & novice discussions targeting business sales & IT departments. Focusing on audio & graphics specific magazines reveals even LESS technical info, and computer technology reviews are often just paid product placements for specific aftermarket OEMs and the like.
Last edited by valis on Wed Dec 31, 2008 8:09 am, edited 1 time in total.
Re: Anandtech's Review Site
Indeed, I totally agree.
Kyle Bennett and Anand were my favorite hardware thrashers when they were partners way back when Computer Shopper Magazine was seen at every RagRack.
KB actually helped my old KeyTech w/ some techno stuff years ago.
His site is also similar now.
But when Kyle, Anand or that sneaky Russian guy @ XBitlabs review something, they leave no stone unturned.
I guess I am sad when I see money corrupting any purely factual process.
THardware does have one useful section w/ certain bench's I favor for longevity & semi audio tests.
I have learned to trust certain people here only as their suggestions and advice have been proven to me over and over with every new modifiaction/build, or additional software purchasing.
Which BTW,.....Thanks Again. Bidule has been upgraded 3 times since I took your advice and I cannot crash that app unless I do ridiculous loads.
I have even watched it load for 15 minutes, massive projects I would never attempt live as the poly I use would devastate them.
But good ole' Bidule takes whatever I throw at it.
My Bassist has the 64bit Vista version doing VST routing chores for his various apps.
Happy New Year Valis.
Kyle Bennett and Anand were my favorite hardware thrashers when they were partners way back when Computer Shopper Magazine was seen at every RagRack.
KB actually helped my old KeyTech w/ some techno stuff years ago.
His site is also similar now.
But when Kyle, Anand or that sneaky Russian guy @ XBitlabs review something, they leave no stone unturned.
I guess I am sad when I see money corrupting any purely factual process.
THardware does have one useful section w/ certain bench's I favor for longevity & semi audio tests.
I have learned to trust certain people here only as their suggestions and advice have been proven to me over and over with every new modifiaction/build, or additional software purchasing.
Which BTW,.....Thanks Again. Bidule has been upgraded 3 times since I took your advice and I cannot crash that app unless I do ridiculous loads.
I have even watched it load for 15 minutes, massive projects I would never attempt live as the poly I use would devastate them.
But good ole' Bidule takes whatever I throw at it.
My Bassist has the 64bit Vista version doing VST routing chores for his various apps.
Happy New Year Valis.
Re: Anandtech's Review Site
Happy New year to you too Jimmy (and anyone else reading this thread). I put in an edit to my post above because I realized I put 'drivers' instead of 'drives' in the first paragraph, potentially confusing.
In my experiences over the years, I've found I don't really need RAID as I don't tax the drives as far as direct throughput goes, but this system I have right now has 4 SATA drives (Seagate ES.2 500Gb w/ 32Mb Cache) and I *still* dislike how the IDE bus can stall a process (especially explorer/desktop) when a drive is being accessed.
I've tried with and without NCQ (AHCI mode) and although NCQ is supposed to replicate SCSI's ability to multitask it's still not really up to snuff. I've been debating moving to an SAS controller using my SATA drives (Seagate's ES.2 drives are their enterprise drives with firmware tuned to be used with SATA/SAS controllers).
This doesn't really make much difference when doing audio work, so it's not something most users here would really care about during musical production. But when I'm doing other tasks (graphics work, 3d stuff, web dev etc) those tiny 'hitches' when attempting to do something while another process is using a drive/sata port still bug me even 13 years after I started using SCSI in my machines. Even Google's Chrome browser (with its multiple processes trying to access the same drive for disk cache) has the ability to cause my computer to stall. 8GB of 667 ddr2 (fb-dimms) with 8 cores (harpertown Xeons), it's the drives that still slow my machine down.
In my experiences over the years, I've found I don't really need RAID as I don't tax the drives as far as direct throughput goes, but this system I have right now has 4 SATA drives (Seagate ES.2 500Gb w/ 32Mb Cache) and I *still* dislike how the IDE bus can stall a process (especially explorer/desktop) when a drive is being accessed.
I've tried with and without NCQ (AHCI mode) and although NCQ is supposed to replicate SCSI's ability to multitask it's still not really up to snuff. I've been debating moving to an SAS controller using my SATA drives (Seagate's ES.2 drives are their enterprise drives with firmware tuned to be used with SATA/SAS controllers).
This doesn't really make much difference when doing audio work, so it's not something most users here would really care about during musical production. But when I'm doing other tasks (graphics work, 3d stuff, web dev etc) those tiny 'hitches' when attempting to do something while another process is using a drive/sata port still bug me even 13 years after I started using SCSI in my machines. Even Google's Chrome browser (with its multiple processes trying to access the same drive for disk cache) has the ability to cause my computer to stall. 8GB of 667 ddr2 (fb-dimms) with 8 cores (harpertown Xeons), it's the drives that still slow my machine down.
Last edited by valis on Wed Dec 31, 2008 8:24 am, edited 1 time in total.
Re: Anandtech's Review Site
Anyway enough about drives, I've observed something more relevant to Jimmy's initial post and the course this thread has taken so far in regards to review sites. It seems to me that in the last 8-9 years the best technical info for hardware in *specific* usages comes not from review sites or technical journals/magazines, but from the users sharing their experiences and collaborating to test the gear. This isn't really anything new, but unlike decades past you have more than a few professionals in a limited area sharing their experiences. I'm thinking of everything from forums like this very one to benchmarks that are community driven and not the result of a publisher or website's dollars (Logic's tests over the years, the DAW benchmark tests, even Cineon's Cinebench tests are done by users with software Cineon is kind enough to contribute).
Re: Anandtech's Review Site
well one more thing about drives, and reviewers.
Make sure you never buy "Seagate 7200.11"
what all the reviewers dont tell you
after they give 10 stars for great speed and 32 meg cache, is that they are made in the factory which used to make "maxtor"
they will die an untimely death either clicking, grinding or one of those other wonderful maxtor traits. they are the first seagate barracuda with only 3 yr warranty instead of 5.
if you want the good non maxtor version get the ES.
Make sure you never buy "Seagate 7200.11"
what all the reviewers dont tell you
after they give 10 stars for great speed and 32 meg cache, is that they are made in the factory which used to make "maxtor"
they will die an untimely death either clicking, grinding or one of those other wonderful maxtor traits. they are the first seagate barracuda with only 3 yr warranty instead of 5.
if you want the good non maxtor version get the ES.
Re: Anandtech's Review Site
This goes slightly off-topic, but
garyb
stardust
valis
Neutron
and
Jimmy V.
Thank you guys, and Happy New Year!
Your advice and wisdom has been the idea I have been led by through these years of my Scope computerizing.
(Sounds maybe corny, but I am even a bit serious)
Happy New Year
garyb
stardust
valis
Neutron
and
Jimmy V.
Thank you guys, and Happy New Year!
Your advice and wisdom has been the idea I have been led by through these years of my Scope computerizing.
(Sounds maybe corny, but I am even a bit serious)
Happy New Year
Re: Anandtech's Review Site
Happy Belated New Year 2U Also.
Here's a nice review from the master himself. They are far and few between, but this guy is really the juice behind reviews and has been benching for years.
We should all love AMD as they make all Intel chipsets cheaper, and forced them to release better designs.
The i7 uses AMD's design of the on die memory controller.
Without the backing of the German Government and AMD, Intel would be raping us.
The new Phenom II's will surely bring down Intels prices by months end.
I love competition.
http://www.anandtech.com/cpuchipsets/sh ... spx?i=3492
Here's a nice review from the master himself. They are far and few between, but this guy is really the juice behind reviews and has been benching for years.
We should all love AMD as they make all Intel chipsets cheaper, and forced them to release better designs.
The i7 uses AMD's design of the on die memory controller.
Without the backing of the German Government and AMD, Intel would be raping us.
The new Phenom II's will surely bring down Intels prices by months end.
I love competition.
http://www.anandtech.com/cpuchipsets/sh ... spx?i=3492
Re: Anandtech's Review Site
Integrating the memory controller in the cpu was inevitable, and not just obvious to AMD & Intel. It's in fact already been done in many low cost 'integrated' type circuits that are less complex, to reduce overall costs of making things out of them.
AMD made a lot of fanfare about it with their press years about about the upcoming "Hammer" cores, because that's what they do. Get big press out of announcing things 6 years before they'll be to market, and keep the attention of tech 'fans' hopefully swaying purchases with promises of things to come. They also do a great job of getting longevity out of a platform, both in terms of cpu socket longevity and in terms of trying to avoid changing requirements too much, obsoleting hardware that can be marketed as being compatible down the line. The recent quiet on their part about newer Phenom parts are due to suffering a backlash of both poorer scaling for their Phenom I design, and the TLB bug that got a lot of press.
Intel does a great job of waiting to release its cards until they absolutely have to, so they can milk a design for all it's worth and still have tech to roll into a process later to yield more performance. I agree though, even though Intel can call AMD's bluff as often as not it's good to have someone around to keep things somewhat honest (even if that brings the inevitable lawsuits eventually). (Incidientally i7 has a TLB bug now too! In fact most cpu's have quite a list of errata when you dig into it, which is usually mitigated by BIOS-level micro-code updates and compilers).
AMD made a lot of fanfare about it with their press years about about the upcoming "Hammer" cores, because that's what they do. Get big press out of announcing things 6 years before they'll be to market, and keep the attention of tech 'fans' hopefully swaying purchases with promises of things to come. They also do a great job of getting longevity out of a platform, both in terms of cpu socket longevity and in terms of trying to avoid changing requirements too much, obsoleting hardware that can be marketed as being compatible down the line. The recent quiet on their part about newer Phenom parts are due to suffering a backlash of both poorer scaling for their Phenom I design, and the TLB bug that got a lot of press.
Intel does a great job of waiting to release its cards until they absolutely have to, so they can milk a design for all it's worth and still have tech to roll into a process later to yield more performance. I agree though, even though Intel can call AMD's bluff as often as not it's good to have someone around to keep things somewhat honest (even if that brings the inevitable lawsuits eventually). (Incidientally i7 has a TLB bug now too! In fact most cpu's have quite a list of errata when you dig into it, which is usually mitigated by BIOS-level micro-code updates and compilers).
Re: Anandtech's Review Site
Brotha' Man Valis,
I have a friend who is asking me for advice on upgrading his studio PC.
He has 2 x PC slave's and MacTel so I told him XITE-1, naturally the price scared him off, so perhaps a cheap Supermicro dual quad seeings how a big price drop is expected soon.
Sure a Nehalem is an option too, but I will working him on certain projects where I can bring in my rack. He has lot's of high end Romplers that seem to work well in 64bit, and the slaved PC's seem to load much less due to their memory addressing limitations.
I am thinking a 64bit PC w/ Dual Xeons would yeild bigger loads and that seems to be the main reason for the upgrade.
Which Xeon chipset and Dual Quads would you suggest?
He is adamant about having quads, even though there's no proof that the NI bundle he bought will benefit.
I know you have experience w/ dual processors so any information would be appreciated.
Actually he's counting on my XITE-1 and Solaris to bolster his synth projects until he gets a grip on Reaktor, etc.
Thanks In Advance.
I have a friend who is asking me for advice on upgrading his studio PC.
He has 2 x PC slave's and MacTel so I told him XITE-1, naturally the price scared him off, so perhaps a cheap Supermicro dual quad seeings how a big price drop is expected soon.
Sure a Nehalem is an option too, but I will working him on certain projects where I can bring in my rack. He has lot's of high end Romplers that seem to work well in 64bit, and the slaved PC's seem to load much less due to their memory addressing limitations.
I am thinking a 64bit PC w/ Dual Xeons would yeild bigger loads and that seems to be the main reason for the upgrade.
Which Xeon chipset and Dual Quads would you suggest?
He is adamant about having quads, even though there's no proof that the NI bundle he bought will benefit.
I know you have experience w/ dual processors so any information would be appreciated.
Actually he's counting on my XITE-1 and Solaris to bolster his synth projects until he gets a grip on Reaktor, etc.
Thanks In Advance.
Re: Anandtech's Review Site
What a Nehelem-generation Xeon will get you
- 4 or 6 cores
- 'enterprise level' prices & hardware availability (platform hasn't even launched yet, Q1 2009 things will 'start' to appear)
- 6 channel ram bus (3 lanes per cpu slot) which is more than any other platform
- motherboards that cost alone as much as entire 'gaming' pc's from Dell/HP/etc
- even more costly ram (since it's not on the market yet either)
- a 4 month wait for parts availability unless you want to pay $1800 per cpu and 2x the cost for ram & motherboard (which will be a rev 1 board with unknown bugs and limitations)
- Even your IT buddies will be impressed (gamers won't even comprehend what the machine is)
What a 5400 series chipset Xeon board will get you:
- Support for insane amounts of ram with insanely hot ram (fb-dimms) that come at a price premium
- Quad channel ram bus (twice the bandwidth of single slot boards, current 5400 series chipsets overcome past issues with memory bandwidth using fb-dimms)
- Support for dual quads (I didn't buy mine for music, I think a single quad of faster speed & faster ram is more cost effective)
- board that costs as much as any elite gamer board
- Much more stringent PSU requirements (PC Power&Cooling is my preference)
- More desire for a dedicated (well ventilated) PC cabinet that has some sound isolation (2 cpu fans instead of one and you'll NEED fans to keep the ram cool)
- Specific case requirements (eATX compatible, with lots of room!)
What a Nehelem core cpu (i7) on a current x58 board will get you:
- First generation Nehelem cpu performance & pricing (slightly more than Penryn, at the bottom of the curve as far as what i7 will yield yet with 'extreme' pricing)
- only 3 choices of cpu speed/price right now
- Typical audio code will be optimized for Nehelem within 9-12 months, at which point the current gen i7 will have surpassed first gen i7 hardware by quite a bit
- Not too many options for heatsinks
- Bragging rights
- You get to find the bugs!
What a decent x48 board with a 3Ghz quadcore Penryn will get you:
- Excellent price:performance ratio for current market offerings
- up to 8GB ram & your choice of PCI & PCIe slots
- ddr2 or ddr3 depending on the board you purchase (ddr3 is more costly but now that we're at 1333mhz fsb & beyond it's starting to show performance gains)
- quieter than a Xeon rig (and cooler!)
- Lower PSU requirements & case requirements
- Noone will diss your PC's gaming abilities, but they won't lust over it either
- Enough power to tackle most things well enough for people who have a workflow honed on PC's from the last decade or so (ie, bouncing tracks can be your FRIEND)
- 4 or 6 cores
- 'enterprise level' prices & hardware availability (platform hasn't even launched yet, Q1 2009 things will 'start' to appear)
- 6 channel ram bus (3 lanes per cpu slot) which is more than any other platform
- motherboards that cost alone as much as entire 'gaming' pc's from Dell/HP/etc
- even more costly ram (since it's not on the market yet either)
- a 4 month wait for parts availability unless you want to pay $1800 per cpu and 2x the cost for ram & motherboard (which will be a rev 1 board with unknown bugs and limitations)
- Even your IT buddies will be impressed (gamers won't even comprehend what the machine is)
What a 5400 series chipset Xeon board will get you:
- Support for insane amounts of ram with insanely hot ram (fb-dimms) that come at a price premium
- Quad channel ram bus (twice the bandwidth of single slot boards, current 5400 series chipsets overcome past issues with memory bandwidth using fb-dimms)
- Support for dual quads (I didn't buy mine for music, I think a single quad of faster speed & faster ram is more cost effective)
- board that costs as much as any elite gamer board
- Much more stringent PSU requirements (PC Power&Cooling is my preference)
- More desire for a dedicated (well ventilated) PC cabinet that has some sound isolation (2 cpu fans instead of one and you'll NEED fans to keep the ram cool)
- Specific case requirements (eATX compatible, with lots of room!)
What a Nehelem core cpu (i7) on a current x58 board will get you:
- First generation Nehelem cpu performance & pricing (slightly more than Penryn, at the bottom of the curve as far as what i7 will yield yet with 'extreme' pricing)
- only 3 choices of cpu speed/price right now
- Typical audio code will be optimized for Nehelem within 9-12 months, at which point the current gen i7 will have surpassed first gen i7 hardware by quite a bit
- Not too many options for heatsinks
- Bragging rights
- You get to find the bugs!
What a decent x48 board with a 3Ghz quadcore Penryn will get you:
- Excellent price:performance ratio for current market offerings
- up to 8GB ram & your choice of PCI & PCIe slots
- ddr2 or ddr3 depending on the board you purchase (ddr3 is more costly but now that we're at 1333mhz fsb & beyond it's starting to show performance gains)
- quieter than a Xeon rig (and cooler!)
- Lower PSU requirements & case requirements
- Noone will diss your PC's gaming abilities, but they won't lust over it either
- Enough power to tackle most things well enough for people who have a workflow honed on PC's from the last decade or so (ie, bouncing tracks can be your FRIEND)
Re: Anandtech's Review Site
Here is my current Xeon board:
http://www.supermicro.com/products/moth ... 7DWA-N.cfm
Price has come down about $70 since I bought it a year ago:
http://www.newegg.com/Product/Product.a ... 6813182130
I have 667mhz ram & 1333cpu's because the 1600mhz capable Xeons for my board still cost 2x as much and are impossible to find. Here's the Xeons @ 3ghz/1333:
Intel Xeon E5450 Harpertown 3.0GHz 12MB L2 Cache LGA 771 80W Quad-Core Processor - $989.99
http://www.newegg.com/Product/Product.a ... 6819117142
Remember that you'd need TWO! (~$1979.98) Now the 3.2Ghz/1600 parts (if you can find them, they get sucked up by HP/Apple/etc usually) will run about $1450-1600 a piece, meaning total outlaw for 2 would be $3000-3200 (which is why I chose 1333mhz parts.)
Here's a board I just used to make a build for a friend:
http://www.supermicro.com/products/moth ... C2SBX+.cfm
Costs half of what my Xeon board did and gives 75-80% of the performance in most audio usages:
http://www.newegg.com/Product/Product.a ... 6813182163
That C2SBX+ could also be paired with 1600mhz capable cpu & ddr3-800 (for 1600mhz fsb operation) MUCH easier than Xeons can.
Ie, for the fastest non-i7 Single slot solution, skip the 'extreme' socket 771 and get this (1600mhz fsb):
Intel Core 2 Extreme QX9770 Yorkfield 3.2GHz 12MB L2 Cache LGA 775 136W Quad-Core Processor - $1,399.99
http://www.newegg.com/Product/Product.a ... 6819115051
However imo THIS is the sweet spot in the price/performance curve for most 'power' users (1333mhz fsb):
Intel Core 2 Quad Q9650 3.0GHz 12MB L2 Cache LGA 775 95W Quad-Core Processor - $549.99
http://www.newegg.com/Product/Product.a ... 6819115130
That Q9650 is what I dropped in the system for him and he's got plenty of horsepower. I didn't compare rompler performance specifically between that machine and my Xeon rig, but overall I'd say that his machine is a bit snappier due to less latency on *each individual channel* of the memory bus (5400 series has 4 ram channels but you don't get parallel bandwidth across all 4 ram busses for single processes, the bandwidth is there for when multiple processes are in play).
Lastly, here are the 3 choices for Nehelem right now, note that the 2 cheaper models are using slower 'QPi' (quick path interconnect) interfaces, ie, slower front-side bus (in legacy terminology):
http://www.newegg.com/Product/ProductLi ... rchInDesc=
i7 still needs time to mature as a platform imo...
http://www.supermicro.com/products/moth ... 7DWA-N.cfm
Price has come down about $70 since I bought it a year ago:
http://www.newegg.com/Product/Product.a ... 6813182130
I have 667mhz ram & 1333cpu's because the 1600mhz capable Xeons for my board still cost 2x as much and are impossible to find. Here's the Xeons @ 3ghz/1333:
Intel Xeon E5450 Harpertown 3.0GHz 12MB L2 Cache LGA 771 80W Quad-Core Processor - $989.99
http://www.newegg.com/Product/Product.a ... 6819117142
Remember that you'd need TWO! (~$1979.98) Now the 3.2Ghz/1600 parts (if you can find them, they get sucked up by HP/Apple/etc usually) will run about $1450-1600 a piece, meaning total outlaw for 2 would be $3000-3200 (which is why I chose 1333mhz parts.)
Here's a board I just used to make a build for a friend:
http://www.supermicro.com/products/moth ... C2SBX+.cfm
Costs half of what my Xeon board did and gives 75-80% of the performance in most audio usages:
http://www.newegg.com/Product/Product.a ... 6813182163
That C2SBX+ could also be paired with 1600mhz capable cpu & ddr3-800 (for 1600mhz fsb operation) MUCH easier than Xeons can.
Ie, for the fastest non-i7 Single slot solution, skip the 'extreme' socket 771 and get this (1600mhz fsb):
Intel Core 2 Extreme QX9770 Yorkfield 3.2GHz 12MB L2 Cache LGA 775 136W Quad-Core Processor - $1,399.99
http://www.newegg.com/Product/Product.a ... 6819115051
However imo THIS is the sweet spot in the price/performance curve for most 'power' users (1333mhz fsb):
Intel Core 2 Quad Q9650 3.0GHz 12MB L2 Cache LGA 775 95W Quad-Core Processor - $549.99
http://www.newegg.com/Product/Product.a ... 6819115130
That Q9650 is what I dropped in the system for him and he's got plenty of horsepower. I didn't compare rompler performance specifically between that machine and my Xeon rig, but overall I'd say that his machine is a bit snappier due to less latency on *each individual channel* of the memory bus (5400 series has 4 ram channels but you don't get parallel bandwidth across all 4 ram busses for single processes, the bandwidth is there for when multiple processes are in play).
Lastly, here are the 3 choices for Nehelem right now, note that the 2 cheaper models are using slower 'QPi' (quick path interconnect) interfaces, ie, slower front-side bus (in legacy terminology):
http://www.newegg.com/Product/ProductLi ... rchInDesc=
i7 still needs time to mature as a platform imo...
Re: Anandtech's Review Site
Hey guys, Anand here. I'm writing this sub-section, not at Gary's request, but because I felt it was necessary. Over the past year I've watched the number of motherboards Gary gets to review go down, and the amount of time spent per motherboard go up tremendously. This year was especially bad as Gary spent more time helping manufacturers fix their BIOSes and compatibility problems than actually writing motherboard reviews.
I wanted to help bring some of what Gary does to light in this section, just so you know the sad state in which many of these motherboards are being brought to market and the work that goes into getting them ready so that we can actually write about it, much less recommend one.
With that said, let's take a look at a particular sequence of events we encountered with the motherboards in today's review. We are not going to name names today as all of the manufacturers are guilty, some worse than others. The point being is that we feel the lack of quality assurance before a product hits the market has now reached an all time high.
______________________________________________________________________________
Nice to see the master and other talent is hard at work.
I wanted to help bring some of what Gary does to light in this section, just so you know the sad state in which many of these motherboards are being brought to market and the work that goes into getting them ready so that we can actually write about it, much less recommend one.
With that said, let's take a look at a particular sequence of events we encountered with the motherboards in today's review. We are not going to name names today as all of the manufacturers are guilty, some worse than others. The point being is that we feel the lack of quality assurance before a product hits the market has now reached an all time high.
______________________________________________________________________________
Nice to see the master and other talent is hard at work.