Recording audio without latency

Hi I have been using Sonar for a while. there is a basic concept that I have not yet understood:
 
When I record vocals, I want to hear my voice without the anoying delay caused by the computer. Further more, I want to hear reverb on my voice as I record.
For the moment, the only way that I have found is to record on a track with the monitoring volume set to zero and have a "pre-fader send" feeding the reverb bus. My dry voice is heard through the audio interface foldback and added to the reverb... After I have recorded a track, if I want to hear it back to do a punch-in, I have to drag my recorded performance onto another track which has the fader to nominal level to hear it in context with reverb, and manually cut it at the desired place... If I do multiple vocal takes I have to drag each them out on different tracks manually and mute them ....I am sure that I am not doing this the right way because it is a slow and clumzy way to work... What am I missing? I work with pro Tools at the studio and this is NEVER an issue, there is no delay in the monitoring and everything is simple... Why isn't there a direct monitoring bus in Sonar that lets you hear the sound directly from the interface when a track is armed and switches automatically to playback when the track is playing? I can't believe there isn't something like this in Sonar... I know there is feature in Cubase and Pro-Tools... When I look at Sonar videos making use of "Record Lanes" and comping within an single track I am frustrated because I can not achieve this in real life...  I refuse to record audio with a 12ms delay in my headphones, I think that in 2015 we have the right to expect not to have to deal with this.. It so happens that when one records vocals, it is often late into the production process, and therefore the session is already heavy with synths and processing, it is then difficult to work at 2ms delay unless you have a $10 000 16 core Xeon.
I expect some of you to suggest to me to print a mix of the backing trak and use it to record the vocals in a lighter session that will enable me to dial down the ASIO buffer, and after that import the vocals back into the main session for mixdown...I think it is a waist of time and besides the point.... 
I saw a Steinberg demo on Youtube that featured an audio inderface that has dsp and enables the use of reverb during recording and there was no fuss about asio delay and muting tracks and moving tracks around like I have to do in Sonar. Everything was integrated and simple. I love my Sonar but I hate recording audio with it...
 
I wish there was a Cackewalk video that explained how one should record audio the most efficient way...How do you guys deal with recording audio?
 
Thank you for your suggestions

I used to get that kind of latency when I had a gammy setup - not that I'm suggesting yours is. Currently though, my mic goes into a ADAT channel mixer/compressor and then into my RME UFX. RME has a mixing software console, which has on-board FX - (basic reverb and echo), but it adds a nice atmosphere when singing...so that's what I listen to in my phones and there's no latency at all. I record this signal directly into Sonar, which can be dry, or I can record the FX from the software if I wanted, but I find I have more control if I just record dry and listen and sing to it wet.

 I refuse to record audio with a 12ms delay in my headphones, I think that in 2015 we have the right to expect not to have to deal with this

 
I hope you don't play guitar.
 
A delay of 12ms is exactly the same as standing 4 metres away from your amp

The very things you say Cubase and Pro Tools were capable of in a video you saw was down to the hardware not the software.

For zero latency monitoring you have to listen to (monitor) directly from your audio interface. If you have the hardware that can do this Sonar is equally capable of zero latency monitoring.

zblip2@gmail.com
Why isn't there a direct monitoring bus in Sonar that lets you hear the sound directly from the interface when a track is armed and switches automatically to playback when the track is playing? I can't believe there isn't something like this in Sonar... I know there is feature in Cubase and Pro-Tools...

As soon as the signal goes inside the box it will be subject to bus (i.e. USB/Firewire/PCIe) transmission latencies, A/D/A conversion and software buffering latencies; there's no avoiding this. The only way you get around it is by direct monitoring where the anlog signal takes an analog path through the interface from input to output with no A/D/A conversion or digital processing.
 
Cubase and Protools support ASIO Direct Monitoring, which is just a way for the software to automatically switch the interface to direct-monitoring mode if the interface also supports it. Not all interfaces do, and because ASIO Direct Monitoring is an extremely loosely/poorly defined "standard" (per Cakewalk CTO Noel Borthwick), Cakewalk have chosen not to implement support for it to date.
 
But that won't solve your problem of wanting to monitor with FX. The only way you get that, other then by having a big, expensive pile of all-analog FX hardware, is by taking the signal into the box, and having hardware that can run plug-in heavy projects with small buffer sizes, and minimal hardware/firmware latencies.
 
You'll also want to avoid tracking with plugins that use internal lookahead buffers that induce Plugin Delay Compensation, although SONAR allows bypassing PDC on input-monitored tracks with live input.

Yes, an interface with onboard FX is another option. You'll still have the A/D/A conversion and whatever buffering the firmware FX require, but that'll be on the order of a 1-2ms, and not troublesome. 

Looks like OP has taken a powder. There's good information here nonetheless.

Define "reasonable price". Give us a budget and the max. number of simultaneous IO required

Bristol_Jonesey
Define "reasonable price". Give us a budget and the max. number of simultaneous IO required

Lets say: 2 mic pres, two audio ins for synths, Stereo out for speakers and another for headphones, an inboard reverb and solid driver?...