Even the fans suck; I can hear it downstairs. The hard drives are SCSI, not even SAS. RAM is like 2 GB, it weighs like 100lb, still has a floppy drive, uses Server 2003, and has an ATI RAGE XL.
Thanks. Asking on behalf of a friend. His landlord who lives next door has offered to share the network for a very low rate. LL has fiber optic and can run a CAT cable and through to a LAN connection in the wall.
Can my friend then simply plug a router into the wall and set up his own secondary network?
@Jdoh Personally when I've got two devices capable of doing NAT, routing and switching, I tend to delegate one device to do all the "Layer 3 stuff" (that is, keeping track of clients on its private subnet) and everybody else just Layer 2 bridges into the Layer 3 network.
He doesn't have a router. He's not particularly tecchy and I'm 5,000km away. I can muddle my way through network stuff but not enough to blindly help him over the phone / remote connection.
@Jdoh not necessarily -- some of the best WiFi infrastructure is "full-stack" with all kinds of enterprisey features, such as the Cisco enterprise kit, that costs $1k and up
I've seen dramatic increases in range, stability, latency, packet loss, jitter, everything -- as the generations have progressed from b to g to n to ac
and 5 GHz is usually way better than 2.4 GHz due to the extreme congestion of 2.4 GHz (Bluetooth, WiFi, microwave oven interference, RF headphones, etc.)
@Bob If I recall correctly, the "big" enhancements for 802.11ac (in terms of top-end throughput, plus beam forming) are strictly unavailable in the 2.4 GHz band.
The only real advantage of 2.4 GHz is, being lower frequency, it can cross thicker physical barriers with less relative attenuation than 5 GHz.
@Jdoh IP range. Uh. Say the landlord has their network on 192.168.1.1/24 (so 192.168.1.1-192.168.1.255). You need to pick a different range. Like 192.168.88/24.
Basically I would recommend buying 802.11ac hardware with all new purchases (of both clients and APs/routers) and make sure it's dual band so you at least have the option to try 5 GHz
@Bob ok, but there are significant reliability improvements in 802.11ac (at least for me) that far surpass 802.11n gear, especially that "Draft N" and "Draft N 2.0" crap that persisted for years in the early 2010s before 802.11n finally stabilized a bit and became less horrible
at this point, 5 GHz 802.11ac is indistinguishable from Ethernet or USB RNDIS for me for general browsing, light downloading and gaming -- even through a floor the speed is quite excellent
@allquicatic I think that also depends a lot on environmental noise.
@Jdoh, you can try to get them to take a look at local noise with a wifi spectrum analyser app on their phone. That'll give you an idea of how congested the 2.4 GHz and 5 GHz bands are. Then you can decide if you need to go up to a higher grade.
@Bob my environmental noise has been very consistent over the years, and I've been using 5 GHz for my main devices since the early Drafts of 802.11n (I even experimented with it on and off with 802.11a when "g" was the latest standard)
basically up and down my suburban neighborhood, almost every house has a cable or ADSL modem, some of them also have one or more routers, and each of these is a wireless base station, and every single one of them is on 2.4 GHz and advertising either 802.11g or 802.11n
I don't contribute to 2.4 GHz noise outside my room because the range on my bluetooth headset is fairly limited (just barely reaches across the hallway)
@Bob my Verizon Jetpack AC791L has a nice feature where it has three extremely user-friendly TX power settings that it lets you choose, and describes the pros and cons of each in very comprehensible terms
(openSUSE Leap gets new versions every twelve months, and each version is supported until two months after the second major release after that version)
My laptop's GPU is running remarkably cool under a Folding@home load.
@bwDraco is folding one of those GPGPU workloads that runs well on both AMD and Nvidia hardware, or is it one of those, like Bitcoin, where a Radeon of almost any scale absolutely smokes even high-end Nvidia chips?
I recall the benchmarks for bitcoin hashes per sec where it was like, Radeon R9 280X several times faster than GTX 980 Ti
supposedly Pascal caught Nvidia GeForce up part-way to AMD in the consumer GPGPU space, but GeForce is still, by far, predominantly a 3d graphics rendering card
whereas Radeon is not particularly more geared towards graphics than GPGPU, at least to the extent that cards that do 3d rendering can be biased or not
(Tesla's still "pure" GPGPU; it can't even render to a framebuffer because it doesn't have a framebuffer)
AMD's been betting on compute for a long time now. That approach has put them at a disadvantage, and game developers have not been doing much GPU compute (aside from PhysX, which is very hard on NVIDIA GPUs) until recently.
AMD's strategy has consistently baffled me. Why waste silicon on compute functionality most consumers will never touch?
NVIDIA sure doesn't.
I mean, if you look at the Radeon HD 7970, FP64 performance was 1/4 of FP32.
If you look at Polaris, they've certainly cut back on compute capability a bit (FP64 at 1/16 of FP32), but they're still using up die space for functionality that >90% of consumers will never touch within the card's lifetime.
NVIDIA understands this. The new FP16 functionality in Pascal is 2x of FP32 in GP100 but a mere 1/64 of FP32 in consumer GPUs.
Polaris, on the other hand, has FP16 at 2x of FP32. 😕
AMD is just wasting die space and power on functionality that will never see use in a consumer environment.
@djsmiley2k It's driving them out of the consumer market, though. Even if their products are viable for heavy GPGPU, most of their raw revenue has to come from sales in laptops and desktops for people who play MMOs, MOBAs and first-person shoooters.
FP16 is useful for AI, and that's what NVIDIA is promoting the GP100 cards for, but has very little usepractically no use in consumer workloads, even ones that use GPGPU functionality.
I can see a select few consumer workloads possibly needing a bit of FP64 in the future, but I see no consumer use case for FP16 (unless you call datacenter workloads "consumer").
really the only consumer GPGPU I see is (1) physics -- which could need FP64 in games like Star Citizen -- and (2) video encoding, which has dedicated silicon for that purpose because dedicated silicon is way faster than general purpose programming and video encoding is too expensive otherwise
@djsmiley2k AMD has been betting on GPGPU for a very long time. The whole concept of the APU was built around the belief that consumer applications will make extensive use of GPGPU functionality in the near future. This hasn't happened.
It's the whole reason they acquired ATI in the first place.
NVIDIA has gimped compute functionality in consumer GPUs not just for market segmentation, but because they rightly foresaw that CUDA functionality would not be extensively used in consumer applications except for a few specialized workloads.
Every bit of silicon that is enabled on the chip wastes power even when it's not used (unless they clock-gate it, but that's besides the point).
As far as I can tell, clock gating in GPUs happens not at individual shader cores, but each cluster of cores (SMs or CUs).
AMD cards wind up wasting more energy on compute resources that are unused 99+% of the time than do NVIDIA cards.
Also, AMD gives more overall die space to compute functionality (including basic FP32 units) than NVIDIA. This is why you see AMD GPUs vastly outperforming otherwise comparable NVIDIA GPUs in certain compute-intensive workloads like cryptocurrency mining, but this hurts their overall gaming performance.
The word Mamihlapinatapai (sometimes spelled mamihlapinatapei) is derived from the Yaghan language of Tierra del Fuego, listed in The Guinness Book of World Records as the "most succinct word", and is considered (for example by Austrian playwright Clemens Berger) one of the hardest words to translate.
== Meaning ==
It allegedly refers to "a look shared by two people, each wishing that the other would initiate something that they both desire but which neither wants to begin." A slightly different interpretation of the meaning also exists: "It is that look across the table when two people are sharing...
It's know that W7 adjusts volume but speaker balance?
And this annoys me!
I have "Loudness Equalization" enabled. Will try to disable but I know volume will overall be weaker. That's why I have this option enabled in 1st place.
Hi everyone. One quick question. I have issue connecting to my server. I can ping it though. I tried to monitor the traffic using wireshark, and I see many tcp retransmissions. What can be the reason?