Recommended Posts

Don't worry yourself over CUDA - ATi and NVIDIA are both committed to supporting OpenCL which is becoming the GPGPU standard (Like OpenGL). It has the benefit of already running on Linux, Windows and Mac OS X and many different companies are backing it including Intel (Who make the majority of GPU's in the Market Place)

That may be the case, but for Intel market share doesn't equate power or quality; if it did I wouldn't own a HD 4870. The only reason I would back Nvidia is because of the the documented progress I've seen from them; it's cool and exciting what the future holds.

If the competition is able to meet or exceed performance at a healthy price then of course all options are on the table.

That may be the case, but for Intel market share doesn't equate power or quality; if it did I wouldn't own a HD 4870. The only reason I would back Nvidia is because of the the documented progress I've seen from them; it's cool and exciting what the future holds.

If the competition is able to meet or exceed performance at a healthy price then of course all options are on the table.

You clearly didn't read my post properly.

Strikes me that you are using your needs to create a standard by which others should build their PCs.

Absolutely not. I didn't bring my needs into this discussion at all, and I'm not sure where you got that idea from.

You want someone to rationalise why anyone could need more than 4GB of RAM?

No, I'm not trying to rationalise a fixed figure and apply it generally to "anyone".

I want anyone to rationalise their need for any amount of RAM with actual numbers reported from actual use. I wanted to see real figures reported by Task Manager, Performance Monitor, etc, but it seems that everybody is perfectly satisfied just saying "take it from me, my app will use that much memory" without any hard numbers to back up those claims. The tools are there to back up the claims, but I guess it's easier to just say "my system feels sluggish if I do such-and-such" and leave it at that.

I am unsure of your computing background or of your job occupation but please take it from me, some of us do indeed require large amounts of RAM and very fast large RAID arrays in order to maintain high data transfer and computation rates. The right tools for the right job.

No doubt some do, I don't doubt that for a nanosecond.

As for my background, I write software for a living, and I've worked with people whose daily job consists of analysing a customer's needs before selling them some overpriced, overkill solution. If anyone of them just went strictly with a gut feeling to put together numbers for a project estimate, they'd lose their job in an instant.

But, I guess there's no point using that same approach here. I guess the approach here is "lets max out the motherboard, and if it runs okay, then I guess it was the right choice", without actually measuring where the point of dimishing returns is. Maybe it was half-way there, but I guess we'll just never know...

He already has 10GB in his Mac Pro and says he needs that much memory. So I don't see the problem. And as I said previously 12GB of ram is ?120 e.g. one of the most cheapest parts of his entire build.

I really think some people on this board take things way to far. He wasn't asking anyone to quantify how much RAM he needed he already knows how much he needs.

Dandy do you realize what your saying see BOZ does stuff more in the lights of what any artist would do in the movies and those programs need a crap tone of ram so BOZ going 12gb of ram is fine and at some point he may max his ram out and as for the GPU situation i would go 3x nvida cards also to the one guy would said something about rendering a 2d window well he may wan to run Windows 7 witch would be running your desktop in DX10 mode 3d in other words

anyways nice looking specs you do want tho bozz

Dandy do you realize what your saying see BOZ does stuff more in the lights of what any artist would do in the movies and those programs need a crap tone of ram so BOZ going 12gb of ram is fine and at some point he may max his ram out and as for the GPU situation i would go 3x nvida cards also to the one guy would said something about rendering a 2d window well he may wan to run Windows 7 witch would be running your desktop in DX10 mode 3d in other words

anyways nice looking specs you do want tho bozz

No punctuation?

Dandy do you realize what your saying see BOZ does stuff more in the lights of what any artist would do in the movies and those programs need a crap tone of ram so BOZ going 12gb of ram is fine and at some point he may max his ram out and as for the GPU situation i would go 3x nvida cards also to the one guy would said something about rendering a 2d window well he may wan to run Windows 7 witch would be running your desktop in DX10 mode 3d in other words

anyways nice looking specs you do want tho bozz

Whilst I totally agree with your sentiments regarding Boz taking the direction he wishes, I have to completely disagree with your remarks on how windows are drawn on Vista and Windows 7.

DWM renders the windows and utilises graphics cards to do so, they are 2D and as such run the graphics card in a 2D mode. Not very resource hungry at all even with dozens or hundreds of windows open. So why you are suggesting Boz would need Tri-SLI to have a lot of windows open is beyond me. As said before the 8800's are fine for that, they are however not fast enough if Boz wishes to do hardware rendering or game.

Boz, all these people telling you that you shouldn't go with 12GB of RAM are a lot of the same ones who told me I shouldn't by a Core i7 965 but to buy a C2D or C2Q instead; I didn't listen to them either!

i would of suggested the 920 instead of a c2d :p 6.4 Q/T is tempting but i couldn't justify the cost over a 920 :p

Strikes me that you are using your needs to create a standard by which others should build their PCs.

You want someone to rationalise why anyone could need more than 4GB of RAM? I do 3D work, and I can tell you right now that the programs I use will eat 4GB of RAM faster than you can load Counter Strike Source. If I start doing particle simulations I would say 12GB of RAM in given situations would actually be tight, thats not even bringing into the equation compositing scenes of 2k or 4K together. I am unsure of your computing background or of your job occupation but please take it from me, some of us do indeed require large amounts of RAM and very fast large RAID arrays in order to maintain high data transfer and computation rates. The right tools for the right job.

work more efficiently :laugh:

i'm rather curious though as great 3d models and (i'm not sure how your particle simulations equates to large amounts of memory usage) have been done in the past with significantly less memory then 12GB?

no doubt someone will find a use for it but could you tell me the extent of the damage what you actually run and use to chew up more then 10GB of ram? or is it just windows using more ram for the sake of using it and you assume its you're own doing?

dandy makes a fairly good argument hes only enquiring to better assist the op in not wasting money its a good thing people try and help others save money (it also seems like his original question was answered aswell)

----------------

To boz:

i don't really know much about the i7 incompatabilities (still need to figure out if they sorted the issue with i7 + gtx 280) but i've heard that inno3d is releasing a gtx 275 with 1792MB of memory i'm not sure if its true or not but the extra memory could certainly be beneficial assuming the price isn't that much higher maybe something to keep an eye out for

i would of suggested the 920 instead of a c2d :p 6.4 Q/T is tempting but i couldn't justify the cost over a 920 :p

work more efficiently :laugh:

i'm rather curious though as great 3d models and (i'm not sure how your particle simulations equates to large amounts of memory usage) have been done in the past with significantly less memory then 12GB?

no doubt someone will find a use for it but could you tell me the extent of the damage what you actually run and use to chew up more then 10GB of ram? or is it just windows using more ram for the sake of using it and you assume its you're own doing?

dandy makes a fairly good argument hes only enquiring to better assist the op in not wasting money its a good thing people try and help others save money (it also seems like his original question was answered aswell)

----------------

To boz:

i don't really know much about the i7 incompatabilities (still need to figure out if they sorted the issue with i7 + gtx 280) but i've heard that inno3d is releasing a gtx 275 with 1792MB of memory i'm not sure if its true or not but the extra memory could certainly be beneficial assuming the price isn't that much higher maybe something to keep an eye out for

Dandy's points are spot on in with regards to apporpriate memory for each task. However there are several factors to consider. Firstly, Boz has been doing his job for a number of years and clearly knows his craft. Using photoshop, for example, to create large poster and banner work requires GBs of RAM, down to the sheer filesizes involved. As for 3D work, sure a 3D model can be done on a small system, but when you are rendering off and compositing data on one machine you need a lot of memory. Secondly RAM prices aren't what they were, better to over estimate slightly in these particular instances than underestimate, since an extra ?60 will be recouped on one project yet the time lost by not having that extra RAM could potentially cost 100x that. Amazes me how so many people are talking down to others when it seems to me like they don't actually have a clue about what their job role entails or requires. Go download a trial version of Realflow from NextLimit, and run a few complex simulations on there, then come back here and tell me how much RAM I/we don't need.

Also, nice way to assume we all solely run Windows:rolleyes::

As for graphics cards EVGA produce two GTX285s one standard and one super clocked, both have 2GB of on board memory.

Edited by ccuk
Dandy's points are spot on in with regards to apporpriate memory for each task. However there are several factors to consider. Firstly, Boz has been doing his job for a number of years and clearly knows his craft. Using photoshop, for example, to create large poster and banner work requires GBs of RAM, down to the sheer filesizes involved. As for 3D work, sure a 3D model can be done on a small system, but when you are rendering off and compositing data on one machine you need a lot of memory.

Fair enough that answers my curiosity :)

Firstly, Boz has been doing his job for a number of years and clearly knows his craft

Firstly not everyone here knows Boz (this is the first time i've heard of him) so not all of us know about boz or how good he is at his job or if he knows his stuff or not thats pure speculation (just like me assuming you use windows :) )

Secondly RAM prices aren't what they were, better to over estimate slightly in these particular instances than underestimate, since an extra ?60 will be recouped on one project yet the time lost by not having that extra RAM could potentially cost 100x that.

i don't disagree with that infact i want to get a large amount of ram for my next build as well my concern is the waste in him buying 12GB(6x2gb) of ram now then a few months down the line 4gb modules come out and then he'll have to decide what to do with the other 6 modules but he may have another system that could use those modules

either way to each his own i'm not complaining about anything my post was purely an inquiry and that i approve of dandy's intentions

Amazes me how so many people are talking down to others when it seems to me like they don't actually have a clue about what their job role entails or requires. Go download a trial version of Realflow from NextLimit, and run a few complex simulations on there, then come back here and tell me how much RAM I/we don't need.

I agree its one thing to voice an opinion and jam it down their throat

Also, nice way to assume we all solely run Windows:rolleyes::

My apologies i thought i read windows somewhere but just noticed the big fat apple in boz's avatar:laugh::

but in any case my question was pointed towards a general issue such as those using vista and assuming they are using as much memory as the os reports they are which influences the results for my question:))

As for graphics cards EVGA produce two GTX285s one standard and one super clocked, both have 2GB of on board memory.

thats great but hes buying 2 gtx 275's not 2 gtx 285's so he may aswell get the one with 2x normal memory (if it exists)

thats great but hes buying 2 gtx 275's not 2 gtx 285's so he may aswell get the one with 2x normal memory (if it exists)

I use Mac OS X, Windows XP, Vista, Server 2003, Ubuntu and recently Fedora.

I understand that, I believe, although I may have misread, he has already purchased the two 275s as they were on special offer. I was merely showing that the GTX285 also comes in a 2GB flavour.

Price wise I would personally love to see a 2GB GTX265 :D If you find one let me know :)

Edited by ccuk
Guys I don't want to sound like an creep but I really didn't ask about advices about ram size. You most likely don't use things as I do. I understand. It's not an overkill when you use 3ds max or photoshop or after effects with hundreds of layers or models. It makes a difference.

You don't sound like a creep to me. I'd say go with 24GB if you can splurge for it. I've had multiple renders going that eats the RAM like there's no tomorrow.

I use Mac OS X, Windows XP, Vista, Server 2003, Ubuntu and recently Fedora.

I understand that, I believe, although I may have misread, he has already purchased the two 275s as they were on special offer. I was merely showing that the GTX285 also comes in a 2GB flavour, I haven't seen a 2GB.

Price wise I would personally love to see a 2GB GTX265 :D If you find one let me know :)

i stick with xp and custom one xp ultimate have used ubuntu a bit but i don't have a need for it yet.

ahh hes already purchased fair enough as for the gtx 265 i'll keep a look ou...wait theres a gtx 265? didn't think that one existed yet :p

You don't sound like a creep to me. I'd say go with 24GB if you can splurge for it. I've had multiple renders going that eats the RAM like there's no tomorrow.

any idea how the tri channel works with 6 dimms? are 3 of the cards just added on for extra storage but don't give any real performance benefits? (2 cards run at half speed to give the same overall performance?)

i've always wondered about that ^_^

In Tri-Channel 3 or 6 Dimms all run together at maximum speed in Tri-Channel mode. Just like when you use Dual-Channel you can use 2 or 4 Dimms

its ashame the full performance can't be utilized since the board maxes out at 25GB/sec and the 920 maxes out at 19.2GB/sec can't wait till 38.4GB/sec is fully supported by mb and processor :D

i don't disagree with that infact i want to get a large amount of ram for my next build as well my concern is the waste in him buying 12GB(6x2gb) of ram now then a few months down the line 4gb modules come out and then he'll have to decide what to do with the other 6 modules but he may have another system that could use those modules

4GB modules of DDR3 are already starting to show up.

Probably allready said but my opinon:

Need SSD for OS/Programs drive. Remnember seek times under 0.01 ms compared 4.7 ms (if have best) or more for HD.

Forget above 920 CPU the 920 clocks well, use extra money SSD Vertex from OCZ.

Mine loads PS 64bit in under 2 seconds, boots under 15 secs. I got 920 @ 4.0Ghz (yes watercooled but air easily beat extreme CPU). 2xVertex RAID 0 dor OS/Programs.

ahh hes already purchased fair enough as for the gtx 265 i'll keep a look ou...wait theres a gtx 265? didn't think that one existed yet :p

I think he has bought them, this thread has turned into a monster thread, so I haven't gone back to check.

I think I got carried away with the 5's there, so a 2GB GTX260 would be nice :D

On the topic of SSD, they are ok, but the performance of the drives degrade over time. So the question comes, are the SSD drives really all that much better than say 2-3 regular SATA II drives in RAID 0 in terms of price/performance, for me at the moment I would say no. Mileage may vary, but for the price of one SSD you can buy 3 hard drives and stick them in RAID 0 and obtain very impressive data throughput and have some money left over for a few beers :D

More power to the original poster. I'm not a Photoshop guy at all, but I do understand that it's known to suck up any amount of RAM you can throw at it. The developer in me can't help but think that its resource management has got to be far from optimal, but then, I'd expect that from any port from a platform utilizing a totally different architecture.

That being said, I'd still love to see a screenshot of Task Manager showing some process details under such heavy loads. Heck, I'd love to hook up a few debugging tools under such circumstances just to see where and how that memory is being used.

In all honesty, I probably brought up that point just to satisfy my own morbid curiosity...

Shoot, I haven't done much graphics recently, but I can still fill my 6gb fairly easily just in games alone! (of course I have lots of other stuff running too <.<)

I have the Asus P6T Deluxe and love it! Running w/ Corsair Dominator ram, and OC'd the i7 920 in 5 minutes to 3.7ghz (probably could go higher, just haven't bothered...cpu hasn't been a bottleneck that I've noticed)

Also, I would suggest Vigor Monsoon III cpu cooler, works great, and fits on the Asus P6T very well (doesn't get in the way of RAM slots like some coolers)

Not sure where you live Boz, but I got the i7 920 at Microcenter for $230!

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Posts

    • Wake me up when this comes to PC. Until then... zzzzzzzz....
    • I was expecting the end of the world to happen before this game or elder scroll 6 to come out.
    • OpenAI and Broadcom unveil Jalapeño, a new AI chip built for LLM inference by Pradeep Viswanathan Image by OpenAI Thanks to the exponential growth of ChatGPT and other LLM-based applications, NVIDIA has grown from a $200 billion company into the first public company to reach a $5 trillion market cap. Even though hyperscalers such as Google and Amazon have their own mature AI accelerators, NVIDIA still dominates the AI infrastructure market with multiple generations of GPUs. Microsoft, OpenAI, and Meta remain among NVIDIA’s largest customers, while Google and Amazon continue to be significant NVIDIA customers as they serve AI workloads for customers on their cloud platforms. Today, OpenAI and Broadcom announced Jalapeño, OpenAI’s first custom “Intelligence Processor” designed specifically for large language model inference. The new chip is the first product from a multi-generation compute platform being developed by OpenAI. OpenAI highlighted that Jalapeño was built from the ground up for current and future LLM workloads, rather than being a general-purpose accelerator adapted for AI. Despite heavy competition from Gemini, Claude, Copilot, and others, ChatGPT remains the most used AI platform in the world. OpenAI mentioned that it leveraged its knowledge of how its models and products run at scale, including ChatGPT, Codex, the API, and future agentic AI systems, to design this new chipset. Its chip architecture reduces data movement while balancing compute, memory, and networking resources. Jalapeño will be deployed in production systems starting in late 2026; however, engineering samples are already running machine learning workloads in OpenAI’s labs at production target frequency and power. According to its internal testing, OpenAI claims this chip can deliver “substantially better” performance per watt, and a detailed technical report is expected in the coming months. While OpenAI designed the chip, Broadcom handled silicon implementation and networking technologies, including Tomahawk networking silicon, and Celestica is assisting with board, rack, and system-level integration. OpenAI pointed out that Jalapeño went from initial design to manufacturing tape-out in just nine months, which it claims is the fastest ASIC development cycle achieved for a high-performance advanced semiconductor. The company attributed the speed of development to its own LLMs, which were used during the chip design and optimization process. Broadcom CEO Hock Tan stated that the company's plan is to deploy the Jalapeño platform at a gigawatt scale with Microsoft and other partners starting in 2026. With Jalapeño, OpenAI joins Google, Microsoft, and Amazon to become a full-stack AI player. The company already develops models and products, and is now moving deeper into infrastructure, including chips, kernels, networking, scheduling, and deployment systems.
    • I'm aware. That information should have been included in the article, making it more complete and information.
    • Converseen 0.15.2.5-2 by Razvan Serea Converseen is a free and open-source batch image converter and resizer. It supports over 100 formats, including DPX, EXR, GIF, JPEG, JPEG-2000, PNG, SVG, TIFF, WebP, HEIC/HEIF, and many others. Users can convert, resize, rotate, flip, and compress multiple images at once. It can also transform entire PDF documents into individual image files. Powered by the ImageMagick library, Converseen features a user-friendly interface and is available in both installer and portable versions. Here’s a list of all the features you can find in Converseen: Batch image conversion (supports 100+ formats) Resize images in bulk Rotate and flip images in bulk Compress images to reduce file size Convert entire PDF documents into image files Support for multiple image formats (JPEG, PNG, TIFF, PDF, BMP, GIF, and more) Customizable output settings (quality, resolution, etc.) Image effects and adjustments (such as brightness, contrast, etc.) Convert images to PDF User-friendly graphical interface Support for drag-and-drop functionality Extract an image from a Windows icon file (*ico) Supports adding watermark to images Portable and installer versions available Leverages ImageMagick for processing power Allows renaming of images in bulk Supports EXIF data editing (for JPEG images) Easy-to-use GUI for non-technical users Command-line support for advanced users Free and open-source software Cross-platform availability Available in multiple languages Download: Converseen 0.15.2.5-2 | Portable | 32-bit | ~40.0 MB (Open Source) View: Converseen Homepage | Screenshot Get alerted to all of our Software updates on Twitter at @NeowinSoftware
  • Recent Achievements

    • First Post
      Tom Schmidt earned a badge
      First Post
    • One Month Later
      D0nn13 earned a badge
      One Month Later
    • Rookie
      +ChiefOfNeo went up a rank
      Rookie
    • One Year In
      Tom Schmidt earned a badge
      One Year In
    • One Month Later
      Tom Schmidt earned a badge
      One Month Later
  • Popular Contributors

    1. 1
      +primortal
      448
    2. 2
      +Edouard
      176
    3. 3
      PsYcHoKiLLa
      123
    4. 4
      Michael Scrip
      81
    5. 5
      Xenon
      75
  • Tell a friend

    Love Neowin? Tell a friend!