|
Hardware Support Discussions related to using various hardware setups with SageTV products. Anything relating to capture cards, remotes, infrared receivers/transmitters, system compatibility or other hardware related problems or suggestions should be posted here. |
|
Thread Tools | Search this Thread | Display Modes |
#21
|
||||
|
||||
Quote:
I was planning on another nvidia graphics card, though its not essential to have the best graphics on this machine as it will only be driving a 20" LCD for the kitchen. Is there something about the intel integrated graphics that I should be looking at? Last edited by Humanzee; 04-08-2008 at 04:17 PM. |
#22
|
||||
|
||||
FWIW, the IP35 pro now has a $70.00 rebate, making it $99.99.
I picked up a Q6600 for $199.99 at microcenter and I can't decide whether to replace my aging sagetv server or my aging workstation (both P4 based machines...). If I go for a new server I will use this mobo. The pci-x slot will go very nicely with my 3ware 9500-S12. If I go workstation I cannot decide whether to go nvidia 790i or intel X48. Of course for $100 the IP35 pro looks very tempting (especially with DDR2 so cheap). It would be nice, for once, to have the latest and greatest but it is tough to swallow $300 for a board and $50+ per gig for DDR3. Decisions, decisions. Jesse
__________________
Server: Asus P5Q-EM DO, Q6600, 8 Gigs ram, WHS 2011, 1 HDHomerun(x2 OTA), 1 HD-PVR, 1 Colossus, V7.1.9 sage, 3.3 TB vid storage. HD100 X1 HD200 X2 HD300 X1 Last edited by Jesse; 04-08-2008 at 04:05 PM. |
#23
|
||||
|
||||
Just realized that my client PC has the intel G33 integrated video. I could possibly move my 8600GTS to the server then but I imagine I still want the extra umph for the main watching / gaming area.
|
#24
|
||||
|
||||
I was just remarking to a co-worker today that the first 'real' computer I owned had a total of 32MB of RAM at $40/8MB. Man things have come a long way. I still have the parts, maybe I could put the old DX50 back together and make a low-power n*x server. It could be over clocked to a whopping 225Mhz.
you guys do have me thinking on what to do for the next server/workstation iteration. dual-core AMD 6400 or OC'd Q6600. Been thinking about OCing this 3200+, but the core is running in the low 50s C already. Dad's new 4000 is in the 30s. I'm almost jealous.
__________________
Server: MS Win7 SP1; FX8350 (H2O cooled); 8GB RAM; Hauppauge HVR-7164 (OTA); HVR-885 (OTA); SageTV 9.1.5.x; 12+TB Sage Storage Clients: HD300 x2; HD200 x2; Placeshifter Service: EPB Fiber (1Gb); OTA (we "cut the cord"); Netflix, Hulu, etc. |
#25
|
|||
|
|||
Quote:
|
#26
|
|||
|
|||
I'd be curious to hear what people have to say about the following concerning how AMD's multi-core implementation is superior to Intel's.
From this thread: http://www.circuitcity.com/ccd/gener...8-9903cd822dd0 comes this post: "Intel and AMD build fundamentally different processors. Intel does not make a true dual core or quad core processor. In the Intel Core Duo, the two cores are on the same die, but completely separate from each other. The two cores can only communicate through the Northbridge chipset (FSB) and despite the shady amrketing by Intelm the core frequency of it's FSB is 400 MHz. This means that the two cores can only communicate with each other and with any other system bus at a max of 400 MHz. Intel realized that this was a huge bottleneck, so they created the Core 2 Duo, which allowed the two cores, still blind to each other, to communicate across the L2 cache. While this was faster than going out to the NB, it introduced a whole new set of problems. The origninal Core 2 Duo's came with very large L2 caches, because the core-to-core comms ate up most of the L2 cache, leaving very little L2 cache available for the system. The longer the computer was on, and the more programs were run, the less L2 cache the system has, and the system begins to crawl. Eventually, Intel figured out a way to address the timing issues with core to core requests, and issued new Core 2 Duos that do not suffer quite as badly as the earlier models. These new C2D's could have less L2 cache and thus were cheaper. They still do not multitask as well as an AMD based system, and they can still communicate with the rest of the system at the speed of the NB/FSB, which is very slow when compared to the speed of the processor. Intel still doesn not build the memory controller on the processor die, nor does it use Hypertransport. These two things are the major reasons why Intel processors, while having faster core frequencies, perform worse in real life than comaprable AMD based systems. Many will counter with synthetic test data that shows that Intel outperforms AMD in real world use, but the testing is misleading. THe Worldbench and SIsmark t ests are designed to the specs of Intel, and are designed to skew data in Intel's favor. It is the nature of Intel being the 800 pound gorilla. The situation gets worse with Intel's Core 2 Quad (?) design, which is tow core 2 duos glued together, but they are independent of each other. Now if core two on die 1 wants to tallk to core 2 on die two, it has to first send the request to die 1 on core 1, which must route the request to the L2 cache, which must send the request to core 1 on die 2, which then send the request to die 2. You can see where this is leading. Intel has announced that it will begin following AMD's lead next year and integrating both the on-die memory controller and the Hypertransport bus. No word yet on whether they intend to make true dual and quad cores (like AMD's) or whether they will continue to build separate cores on the same die, which for Intel is cheaper and reduces the number of bad chips."
__________________
STV 9.1.5.683 / Vista Premium / Compaq Presario SR-5550F / AMD Athlon 64 X2 5400 / nvidia GeForce 8400GS / 3 GB DDR2 / 500 GB SATA / 1 TB SATA / M2N68-LA Motherboard / Hauppauge HVR-1250 / Hauppauge HVR-1600 / 2 x STX-HD100 / OTA / Channel Master FLATenna 35 |
#27
|
|||
|
|||
When they start making boards with Intel northbridges and AMD sockets I wouldn't bother me. My troubles are with the northbridges.
|
#28
|
||||
|
||||
Quote:
He's outright wrong on the FSB speeds, even putting the misleading AMD spin on it. Intel FSBs are "quad-pumped" (marketing term). That means that data is being transfered 4 times per physical clock cycle. So while it is technically accurate to say that the FSB's clock is 266MHz or 333MHz, the actual bandwidth is equivalent to a 1066MHz or 1333MHz FSB (respectively) with a single data transfer per clock. I don't know where he pulled the 400MHz number, unless he was talking about the fairly new 1600MHZ FSB QX9770, which has a 400MHz clock on the FSB, quad-pumped to 1600MHz. I guess he could have been talking about overclocking... A bunch of the other bits are more or less correct as far as how the chips are put together. AMD is definitely a "4-core" chip, while Intel is 2 2-core chips... But the part where it really goes off into fanboi-ism is regarding performance. Clock-for-clock, in real-world applications (encoding, games, etc...) Intel easily beats AMD. In the end, Intel has the performance advantage (especially with the new 45nm chips), but it comes at a higher cost. Intel also has a pretty significant lead in encoding speeds, which will jump more when SSE4 goes into common usage. AMD can't match the performance, so they are having to compete on price. You'll get a better price/performance ratio with AMD, so if your application isn't bleeding-edge AMD works well. Lastly, overclocking on the Intel CPUs is crazy, and only getting moreso with the 45nm chips. With the right motherboard and RAM It's fairly easy to take a $200 2.4GHz Q6600 up to 3-3.2GHz on air cooling. The 2.6GHz QX9450s are $310, and can get up to 3.6GHz on air... Quote:
__________________
SageTV V7 (WHS), Diamond UI Server: WHS with Xeon X3350, 4GB ECC, ASUS P5BV-C/4L, recording into a 6.6TB Drive pool Tuners: 4 (2x HDHR) Clients: 2x HD300, 1x HD200 Extenders, 1x Placeshifter 2x Roku XD |
#29
|
||||
|
||||
Quote:
Ha! we have come a long way...
__________________
Eckwell / Chicago, IL SERVER: AMD Phenom III 8750 Black / 4 GB / XP Pro / JV1.6.24 / V7 / 2 PVR-HD (DirectTV) & HDHR (2xATSC) STORAGE: WD Raptor36GB boot / 11TB in 2 eSATA enclosure DEFAULT STV Comskip triggered by DirMON2 --- HD100 / HD200 /HD300 BMI fanart |
#30
|
||||
|
||||
not trying to hy-jack the thread
Quote:
Maybe all this DC/QC stuff will get sorted out by the time the wife 'allows' me to upgrade the server.
__________________
Server: MS Win7 SP1; FX8350 (H2O cooled); 8GB RAM; Hauppauge HVR-7164 (OTA); HVR-885 (OTA); SageTV 9.1.5.x; 12+TB Sage Storage Clients: HD300 x2; HD200 x2; Placeshifter Service: EPB Fiber (1Gb); OTA (we "cut the cord"); Netflix, Hulu, etc. |
#31
|
|||
|
|||
Quote:
My motherboard was the DFI Lanparty UT nF3 250GB, an otherwise superb motherboard. I ended up disabling the onboard ethernet and putting in an Intel PCI NIC card. Slower.....but the problems with streaming video went away. -Robert Last edited by valnar; 04-10-2008 at 12:08 PM. |
#32
|
||||
|
||||
I think I've decided now on the IP35 Pro with a Q6600 thanks to CollinR's persuasion. Ill have to get a serial card, or a usb to serial adapter. That and maybe a non realtek nic for my client PC.
|
#33
|
|||
|
|||
Quote:
I can't speak to the earlier Marvell, but I understand the more recent Marvell LAN chipset to be superior to the Realtek. Video streaming on a gigabit network should not really care about the brand of chipset, too low a bandwidth hit. I would suspect a problem with the nForce 3 chips. |
#34
|
|||
|
|||
Quote:
Robert |
Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
AMD cpu & a Windows timing bug affects SageTV | Opus4 | Hardware Support | 30 | 03-11-2012 07:36 PM |
SageTV Announces Support for AMD LIVE!(TM) Home Cinema and Home Media Server | Narflex | Announcements | 0 | 01-10-2007 09:09 AM |
Intel or AMD for SageTV PC? | CadErik | Hardware Support | 6 | 11-20-2006 07:03 PM |
Problems transcoding with AMD processor? | popechild | SageTV Beta Test Software | 9 | 10-24-2006 05:49 PM |
Intel D101GGC Mobo | Jesse | Hardware Support | 3 | 04-27-2006 03:24 PM |